Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

.hfignore

@@ -0,0 +1,3 @@
+checkpoint-*
+checkpoint-**
+checkpoint-**/**

README.md

@@ -1,3 +0,0 @@
----
-license: cc-by-nc-nd-4.0
----

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 35.0,
+    "eval_loss": 2.5913538932800293,
+    "eval_runtime": 0.4619,
+    "eval_samples_per_second": 303.127,
+    "eval_steps_per_second": 2.165,
+    "total_flos": 1.12061156850611e+18,
+    "train_loss": 0.2908882966189056,
+    "train_runtime": 4060.664,
+    "train_samples_per_second": 119.463,
+    "train_steps_per_second": 0.241
+}

chat_template.jinja

@@ -0,0 +1 @@
+{% for message in messages %}{% if message['role'] == 'user' %}{{'<用户>' + message['content'] + '<AI>'}}{% else %}{{message['content']}}{% endif %}{% endfor %}

config.json

@@ -0,0 +1,36 @@
+{
+  "architectures": [
+    "MiniCPMForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_minicpm.MiniCPMConfig",
+    "AutoModel": "modeling_minicpm.MiniCPMForCausalLM",
+    "AutoModelForCausalLM": "modeling_minicpm.MiniCPMForCausalLM",
+    "AutoModelForSeq2SeqLM": "modeling_minicpm.MiniCPMForCausalLM",
+    "AutoModelForSequenceClassification": "modeling_minicpm.MiniCPMForSequenceClassification"
+  },
+  "bos_token_id": 1,
+  "dim_model_base": 256,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2304,
+  "initializer_range": 0.1,
+  "intermediate_size": 5760,
+  "max_position_embeddings": 4096,
+  "model_type": "minicpm",
+  "num_attention_heads": 36,
+  "num_hidden_layers": 40,
+  "num_key_value_heads": 36,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "scale_depth": 1.4,
+  "scale_emb": 12,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.52.1",
+  "use_cache": false,
+  "vocab_size": 122753
+}

configuration_minicpm.py

@@ -0,0 +1,202 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" MiniCPM model configuration"""
+
+from transformers.configuration_utils import PretrainedConfig
+from transformers.utils import logging
+
+
+logger = logging.get_logger(__name__)
+
+MINICPM_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
+
+
+class MiniCPMConfig(PretrainedConfig):
+    r"""
+    This is the configuration class to store the configuration of a [`MiniCPMModel`]. It is used to instantiate an MiniCPM
+    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
+    defaults will yield a similar configuration to that of the MiniCPM-7B.
+
+    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
+    documentation from [`PretrainedConfig`] for more information.
+
+
+    Args:
+        vocab_size (`int`, *optional*, defaults to 32000):
+            Vocabulary size of the MiniCPM model. Defines the number of different tokens that can be represented by the
+            `inputs_ids` passed when calling [`MiniCPMModel`]
+        hidden_size (`int`, *optional*, defaults to 4096):
+            Dimension of the hidden representations.
+        intermediate_size (`int`, *optional*, defaults to 11008):
+            Dimension of the MLP representations.
+        num_hidden_layers (`int`, *optional*, defaults to 32):
+            Number of hidden layers in the Transformer decoder.
+        num_attention_heads (`int`, *optional*, defaults to 32):
+            Number of attention heads for each attention layer in the Transformer decoder.
+        num_key_value_heads (`int`, *optional*):
+            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
+            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
+            `num_key_value_heads=1 the model will use Multi Query Attention (MQA) otherwise GQA is used. When
+            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
+            by meanpooling all the original heads within that group. For more details checkout [this
+            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
+            `num_attention_heads`.
+        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
+            The non-linear activation function (function or string) in the decoder.
+        max_position_embeddings (`int`, *optional*, defaults to 2048):
+            The maximum sequence length that this model might ever be used with. MiniCPM 1 supports up to 2048 tokens,
+            MiniCPM 2 up to 4096, CodeMiniCPM up to 16384.
+        initializer_range (`float`, *optional*, defaults to 0.02):
+            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
+        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
+            The epsilon used by the rms normalization layers.
+        use_cache (`bool`, *optional*, defaults to `True`):
+            Whether or not the model should return the last key/values attentions (not used by all models). Only
+            relevant if `config.is_decoder=True`.
+        pad_token_id (`int`, *optional*):
+            Padding token id.
+        bos_token_id (`int`, *optional*, defaults to 1):
+            Beginning of stream token id.
+        eos_token_id (`int`, *optional*, defaults to 2):
+            End of stream token id.
+        pretraining_tp (`int`, *optional*, defaults to 1):
+            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
+            document](https://huggingface.co/docs/transformers/parallelism) to understand more about it. This value is
+            necessary to ensure exact reproducibility of the pretraining results. Please refer to [this
+            issue](https://github.com/pytorch/pytorch/issues/76232).
+        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
+            Whether to tie weight embeddings
+        rope_theta (`float`, *optional*, defaults to 10000.0):
+            The base period of the RoPE embeddings.
+        rope_scaling (`Dict`, *optional*):
+            Dictionary containing the scaling configuration for the RoPE embeddings. Currently supports two scaling
+            strategies: linear and dynamic. Their scaling factor must be a float greater than 1. The expected format is
+            `{"type": strategy name, "factor": scaling factor}`. When using this flag, don't update
+            `max_position_embeddings` to the expected new maximum. See the following thread for more information on how
+            these scaling strategies behave:
+            https://www.reddit.com/r/LocalMiniCPM/comments/14mrgpr/dynamically_scaled_rope_further_increases/. This is an
+            experimental feature, subject to breaking API changes in future versions.
+        attention_bias (`bool`, defaults to `False`, *optional*, defaults to `False`):
+            Whether to use a bias in the query, key, value and output projection layers during self-attention.
+        attention_dropout (`float`, *optional*, defaults to 0.0):
+            The dropout ratio for the attention probabilities.
+
+    ```python
+    >>> from transformers import MiniCPMModel, MiniCPMConfig
+
+    >>> # Initializing a MiniCPM minicpm-7b style configuration
+    >>> configuration = MiniCPMConfig()
+
+    >>> # Initializing a model from the minicpm-7b style configuration
+    >>> model = MiniCPMModel(configuration)
+
+    >>> # Accessing the model configuration
+    >>> configuration = model.config
+    ```"""
+
+    model_type = "minicpm"
+    keys_to_ignore_at_inference = ["past_key_values"]
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        hidden_size=4096,
+        intermediate_size=11008,
+        num_hidden_layers=32,
+        num_attention_heads=32,
+        num_key_value_heads=None,
+        hidden_act="silu",
+        max_position_embeddings=2048,
+        initializer_range=0.02,
+        rms_norm_eps=1e-6,
+        use_cache=True,
+        pad_token_id=None,
+        bos_token_id=1,
+        eos_token_id=2,
+        pretraining_tp=1,
+        tie_word_embeddings=True,
+        rope_theta=10000.0,
+        rope_scaling=None,
+        attention_bias=False,
+        attention_dropout=0.0,
+        scale_emb=1,
+        dim_model_base=1,
+        scale_depth=1,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.max_position_embeddings = max_position_embeddings
+        self.hidden_size = hidden_size
+        self.intermediate_size = intermediate_size
+        self.num_hidden_layers = num_hidden_layers
+        self.num_attention_heads = num_attention_heads
+
+        # for backward compatibility
+        if num_key_value_heads is None:
+            num_key_value_heads = num_attention_heads
+
+        self.num_key_value_heads = num_key_value_heads
+        self.hidden_act = hidden_act
+        self.initializer_range = initializer_range
+        self.rms_norm_eps = rms_norm_eps
+        self.pretraining_tp = pretraining_tp
+        self.use_cache = use_cache
+        self.rope_theta = rope_theta
+        self.rope_scaling = rope_scaling
+        self._rope_scaling_validation()
+        self.attention_bias = attention_bias
+        self.attention_dropout = attention_dropout
+        self.scale_emb = scale_emb
+        self.dim_model_base = dim_model_base
+        self.scale_depth = scale_depth
+
+        super().__init__(
+            pad_token_id=pad_token_id,
+            bos_token_id=bos_token_id,
+            eos_token_id=eos_token_id,
+            tie_word_embeddings=tie_word_embeddings,
+            **kwargs,
+        )
+        try:
+            import flash_attn
+            self._attn_implementation = "flash_attention_2"
+        except:
+            pass
+
+    def _rope_scaling_validation(self):
+        """
+        Validate the `rope_scaling` configuration.
+        """
+        if self.rope_scaling is None:
+            return
+
+        if not isinstance(self.rope_scaling, dict) or len(self.rope_scaling) != 2:
+            raise ValueError(
+                "`rope_scaling` must be a dictionary with with two fields, `type` and `factor`, "
+                f"got {self.rope_scaling}"
+            )
+        rope_scaling_type = self.rope_scaling.get("type", None)
+        rope_scaling_factor = self.rope_scaling.get("factor", None)
+        if rope_scaling_type is None or rope_scaling_type not in ["linear", "dynamic"]:
+            raise ValueError(
+                f"`rope_scaling`'s type field must be one of ['linear', 'dynamic'], got {rope_scaling_type}"
+            )
+        if rope_scaling_factor is None or not isinstance(rope_scaling_factor, float) or rope_scaling_factor <= 1.0:
+            raise ValueError(f"`rope_scaling`'s factor field must be a float > 1, got {rope_scaling_factor}")

eval_gen/gpu_results/results_gpu_0.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone is stroking the belly of an animal ⇔ ", "response": "The woman is stroking the animal's belly", "predicted": -100, "true_label": "A woman is stroking a lemur"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is stroking a lemur ⇔ ", "response": "A woman is holding a lemur", "predicted": -100, "true_label": "Someone is stroking the belly of an animal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman using an eye pencil and applying eye liner to her eyelid ⇔ ", "response": "A woman is applying cosmetics to her eyelid", "predicted": -100, "true_label": "A woman is using an eye pencil and is applying eye liner to her eyelid"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is using an eye pencil and is applying eye liner to her eyelid ⇔ ", "response": "A woman is applying cosmetics to her eyelid", "predicted": -100, "true_label": "There is no woman using an eye pencil and applying eye liner to her eyelid"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我爱你．粤语怎么说．？ ⇔ ", "response": "我想你．粤语怎么说．？", "predicted": -100, "true_label": "我爱你用粤语怎么说"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我爱你用粤语怎么说 ⇔ ", "response": "我爱你用粤语怎么说", "predicted": -100, "true_label": "我爱你．粤语怎么说．？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is slicing a tomato ⇔ ", "response": "There is no man slicing a tomato", "predicted": -100, "true_label": "There is no man slicing a tomato"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man slicing a tomato ⇔ ", "response": "A man is slicing a tomato", "predicted": -100, "true_label": "A man is slicing a tomato"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn April , it had forecast operating earnings in the range of 60 to 80 cents a share . ⇔ ", "response": "It had previously forecast second-quarter operating earnings between 60 cents and 80 cents per share .", "predicted": 60, "true_label": "Kodak expects earnings of 5 cents to 25 cents a share in the quarter ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nKodak expects earnings of 5 cents to 25 cents a share in the quarter . ⇔ ", "response": "Kodak also expects to report a quarter profit between 5 and 25 cents a share .", "predicted": 5, "true_label": "In April , it had forecast operating earnings in the range of 60 to 80 cents a share ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoA5root ⇔ ", "response": "OPPOA5ota", "predicted": 5, "true_label": "oppor9ota"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor9ota ⇔ ", "response": "opport9m", "predicted": 9, "true_label": "oppoA5root"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了饭吗？ ⇔ ", "response": "我今天天吃了饭？", "predicted": -100, "true_label": "我今天吃了饭马？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了饭马？ ⇔ ", "response": "我今天吃了饭吗？", "predicted": -100, "true_label": "我今天吃了饭吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man wearing white turns to hit a yellow tennis ball. ⇔ ", "response": "A man is hitting a ball.", "predicted": -100, "true_label": "Someone takes a swing at a ball."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone takes a swing at a ball. ⇔ ", "response": "A baseball player is hitting the ball", "predicted": -100, "true_label": "A man wearing white turns to hit a yellow tennis ball."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n啤酒鸭怎么做 ⇔ ", "response": "烤鸭怎么吃", "predicted": -100, "true_label": "鸭鸭怎么叫"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n鸭鸭怎么叫 ⇔ ", "response": "鸭叫吗", "predicted": -100, "true_label": "啤酒鸭怎么做"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA9osie ⇔ ", "response": "OPPOr9", "predicted": 9, "true_label": "OSIE"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOSIE ⇔ ", "response": "esim", "predicted": -100, "true_label": "A9osie"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n同样地，如果满足其他条件，皈依伊斯兰教的非马来马来西亚人可以声称拥有土着特权。 ⇔ ", "response": "同样地，如果满足其他条件，皈依伊斯兰教的非马来马来西亚人可以声称拥有土着特权。", "predicted": -100, "true_label": "同样地，一名皈依伊斯兰教的非马来马来西亚人可以申请土着特权，只要他符合其他条件。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n同样地，一名皈依伊斯兰教的非马来马来西亚人可以申请土着特权，只要他符合其他条件。 ⇔ ", "response": "一名皈依伊斯兰教的非马来马来西亚人也可以申请土著特权，只要他符合其他条件。", "predicted": -100, "true_label": "同样地，如果满足其他条件，皈依伊斯兰教的非马来马来西亚人可以声称拥有土着特权。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好 ⇔ ", "response": "今天天气真123", "predicted": 123, "true_label": "今天天气真郁"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真郁 ⇔ ", "response": "今天天气真好", "predicted": -100, "true_label": "今天天气真好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nhhdhhd ⇔ ", "response": "hhhdd", "predicted": -100, "true_label": "lhdc"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nlhdc ⇔ ", "response": "hghh", "predicted": -100, "true_label": "hhdhhd"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我耍流氓去了。 ⇔ ", "response": "我去玩街头篮球了。", "predicted": -100, "true_label": "我耍流氓去了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我耍流氓去了 ⇔ ", "response": "我去耍流氓了", "predicted": -100, "true_label": "我耍流氓去了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is awakening ⇔ ", "response": "A girl is sleeping", "predicted": -100, "true_label": "A child is waking up"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child is waking up ⇔ ", "response": "A child is falling back to sleep in bed", "predicted": -100, "true_label": "A girl is awakening"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoace2 ⇔ ", "response": "OPPO Ace2", "predicted": 2, "true_label": "opporoneace2a"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nopporoneace2a ⇔ ", "response": "OPPOAce2", "predicted": 2, "true_label": "oppoace2"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n尽管许多催化应用中的铁通常活性较低，但它比其他金属更便宜且更环保。 ⇔ ", "response": "相对于其他金属，尽管许多催化应用中的铁通常活性较低，但它更便宜且更环保。", "predicted": -100, "true_label": "尽管许多催化应用中的铁通常较便宜，但它的活性较低，并且比其他金属更环保。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n尽管许多催化应用中的铁通常较便宜，但它的活性较低，并且比其他金属更环保。 ⇔ ", "response": "尽管许多催化应用中的铁通常较便宜，但它的活性较低，并且比其他金属更环保。", "predicted": -100, "true_label": "尽管许多催化应用中的铁通常活性较低，但它比其他金属更便宜且更环保。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样赚话费 ⇔ ", "response": "如何赚话费", "predicted": -100, "true_label": "怎样赚话费？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样赚话费？ ⇔ ", "response": "话费充值中心", "predicted": -100, "true_label": "怎样赚话费"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我这样的电脑可以玩什么游戏 ⇔ ", "response": "我这样的电脑可以玩什么软件", "predicted": -100, "true_label": "我的烂电脑可以玩什么游戏"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的烂电脑可以玩什么游戏 ⇔ ", "response": "能吃吗游戏怎么玩", "predicted": -100, "true_label": "我这样的电脑可以玩什么游戏"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCadbury Schweppes plc plans to cut 5500 jobs and shut factories after a 4.9 billion ( $ A11.9 billion ) acquisition spree over the past three years inflated costs . ⇔ ", "response": "London 's Cadbury Schweppes said yesterday it would cut 5500 jobs and shut factories after a 4.9 billion dollar ( $ 11.9 billion ) acquisition spree inflated costs .", "predicted": 5500, "true_label": "Cadbury Schweppes has unveiled plans to slash 5,500 jobs and 20 percent of its factories over four years to cut costs brought about by an acquisition spree ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCadbury Schweppes has unveiled plans to slash 5,500 jobs and 20 percent of its factories over four years to cut costs brought about by an acquisition spree . ⇔ ", "response": "The drinks firm Cadbury Schweppes said it would cut 5,500 jobs and 20 per cent of its factories over four years to compensate for cost rises brought about by its acquisition spree .", "predicted": 5, "true_label": "Cadbury Schweppes plc plans to cut 5500 jobs and shut factories after a 4.9 billion ( $ A11.9 billion ) acquisition spree over the past three years inflated costs ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAfter a seven-day trial last year , Thompson found the monument to be an unconstitutional endorsement of religion by the state . ⇔ ", "response": "Thompson argued the monument was an unconstitutional endorsement of religion by the state .", "predicted": -100, "true_label": "Last year , Thompson ruled that the monument was an unconstitutional endorsement of religion by the state ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nLast year , Thompson ruled that the monument was an unconstitutional endorsement of religion by the state . ⇔ ", "response": "Last year Mr. Thompson ruled that the monument was an unconstitutional endorsement of religion by the state .", "predicted": -100, "true_label": "After a seven-day trial last year , Thompson found the monument to be an unconstitutional endorsement of religion by the state ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n青岛的风湿医院有哪些 ⇔ ", "response": "郑州哪家医院看风湿最好", "predicted": -100, "true_label": "青岛有哪些好的风湿医院"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n青岛有哪些好的风湿医院 ⇔ ", "response": "青岛有哪些好的风湿免疫医院", "predicted": -100, "true_label": "青岛的风湿医院有哪些"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a black tank top wearing a red plaid hat ⇔ ", "response": "A man in a black tank top is wearing a red plaid hat.", "predicted": -100, "true_label": "A man in a hat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a hat. ⇔ ", "response": "A man wearing a black hat stands in the doorway with another man wearing a blue hat.", "predicted": -100, "true_label": "A man in a black tank top wearing a red plaid hat"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nfhdm ⇔ ", "response": "mhft", "predicted": -100, "true_label": "jfjghchdbbffjkill"}
+{"prompt": "Please generate a paraphrase for the following sentence.\njfjghchdbbffjkill ⇔ ", "response": "fhdc", "predicted": -100, "true_label": "fhdm"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree young men wearing backpacks, shorts, t-shirts, sandals and hats are standing together on a sandy surface with palm trees in the distance. ⇔ ", "response": "The three men are outdoors.", "predicted": -100, "true_label": "A small group of men clad in summery clothing are standing on sandy palm tree lined beach."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA small group of men clad in summery clothing are standing on sandy palm tree lined beach. ⇔ ", "response": "A group of people is on a beach.", "predicted": -100, "true_label": "Three young men wearing backpacks, shorts, t-shirts, sandals and hats are standing together on a sandy surface with palm trees in the distance."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n叶酸什么牌子好 ⇔ ", "response": "叶酸什么品牌好", "predicted": -100, "true_label": "叶酸什么牌子好多少钱"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n叶酸什么牌子好多少钱 ⇔ ", "response": "叶酸什么品牌好？多少钱？", "predicted": -100, "true_label": "叶酸什么牌子好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n持刀男子在白金汉宫入口被捕 ⇔ ", "response": "持刀男子被抓获进入白金汉宫", "predicted": -100, "true_label": "白金汉宫被捕男子被落案起诉"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n白金汉宫被捕男子被落案起诉 ⇔ ", "response": "被捕男子逃犯白金汉宫落案起诉", "predicted": -100, "true_label": "持刀男子在白金汉宫入口被捕"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople, mainly women and 2 referees, are skating fiercely during a roller derby match. ⇔ ", "response": "There are people skating.", "predicted": -100, "true_label": "Fierce competition in the roller derby women's final match ensues."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFierce competition in the roller derby women's final match ensues. ⇔ ", "response": "The women's roller derby final is in full swing with fierce competition.", "predicted": -100, "true_label": "People, mainly women and 2 referees, are skating fiercely during a roller derby match."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他去超市买了一些苹果。 ⇔ ", "response": "他去超市买了一些水果。", "predicted": -100, "true_label": "他去超市买了写苹果。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他去超市买了写苹果。 ⇔ ", "response": "他去超市买了苹果。", "predicted": -100, "true_label": "他去超市买了一些苹果。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man dressed in black and white is standing on the side of the street eating an ice cream cone. ⇔ ", "response": "A man is standing on the street.", "predicted": -100, "true_label": "The man is eating chocolate ice cream."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is eating chocolate ice cream. ⇔ ", "response": "A young man eating while wearing a t-shirt and jeans.", "predicted": -100, "true_label": "A man dressed in black and white is standing on the side of the street eating an ice cream cone."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气很漂亮，适合出去玩。 ⇔ ", "response": "今天天气很漂亮，适合去爪。", "predicted": -100, "true_label": "今天天气很漂浪，适合出去玩。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气很漂浪，适合出去玩。 ⇔ ", "response": "今天天气很好，适合出去玩。", "predicted": -100, "true_label": "今天天气很漂亮，适合出去玩。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOA7 ⇔ ", "response": "oppooppoa83", "predicted": 83, "true_label": "opoopa73"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nopoopa73 ⇔ ", "response": "Oppoa8", "predicted": 8, "true_label": "OPPOA7"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA crowd of people shopping at a street market in an urban area with buildings and a statue in background. ⇔ ", "response": "A group of people are outside.", "predicted": -100, "true_label": "the crowd of people are shopping."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nthe crowd of people are shopping. ⇔ ", "response": "People are crowded together in a mall.", "predicted": -100, "true_label": "A crowd of people shopping at a street market in an urban area with buildings and a statue in background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is brushing her hair ⇔ ", "response": "There is no girl brushing her hair", "predicted": -100, "true_label": "A girl is styling her hair"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is styling her hair ⇔ ", "response": "There is no girl styling her hair", "predicted": -100, "true_label": "A girl is brushing her hair"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么睡觉时会打呼噜 ⇔ ", "response": "为什么睡觉时会打哈欠", "predicted": -100, "true_label": "现在疫情怎么样"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n现在疫情怎么样 ⇔ ", "response": "现在上海疫情怎么样？", "predicted": -100, "true_label": "为什么睡觉时会打呼噜"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" If you pass this bill , Big Brother will be watching you , \" said Rep. John Mabry , D-Waco . ⇔ ", "response": "\" If you pass this bill, Big Brother will be watching you , \" 150-seat conference room of the Gruene Hall in Gruene, located in the heart of the Hill Country , was filled with .", "predicted": 150, "true_label": "\" If you pass this bill , \" Rep. John Mabry Jr . , D-Waco , told colleagues , \" Big Brother will be watching you . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" If you pass this bill , \" Rep. John Mabry Jr . , D-Waco , told colleagues , \" Big Brother will be watching you . \" ⇔ ", "response": "\" If you pass this bill, Congressman John Mabry told his colleagues, ' Big Brother will be watching you. ' \"", "predicted": -100, "true_label": "\" If you pass this bill , Big Brother will be watching you , \" said Rep. John Mabry , D-Waco ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n啊？彼烈霍多夫是自作自受，不错把公款拿去挥霍开玩笑！ ⇔ ", "response": "哦，是的，是彼烈霍多夫自作自受，把公款拿去挥霍开开玩笑！", "predicted": -100, "true_label": "啊？彼列霍多夫的确是自作自受挪用了公款开玩笑！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n啊？彼列霍多夫的确是自作自受挪用了公款开玩笑！ ⇔ ", "response": "啊？列霍多夫自作自受挪用了公款当笑话开！", "predicted": -100, "true_label": "啊？彼烈霍多夫是自作自受，不错把公款拿去挥霍开玩笑！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is making a sketch ⇔ ", "response": "A man is sketching a scene", "predicted": -100, "true_label": "A man is drawing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is drawing ⇔ ", "response": "There is no man drawing", "predicted": -100, "true_label": "A man is making a sketch"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is doing back flips on a trampoline ⇔ ", "response": "A man is doing flips off of a trampoline", "predicted": -100, "true_label": "A man is jumping on a trampoline"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is jumping on a trampoline ⇔ ", "response": "A man is jumping indoors", "predicted": -100, "true_label": "A man is doing back flips on a trampoline"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n给我听一些笑话 ⇔ ", "response": "讲个笑话吧", "predicted": -100, "true_label": "讲个笑话的声音"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n讲个笑话的声音 ⇔ ", "response": "你愿意我帮你倒杯水吗", "predicted": -100, "true_label": "给我听一些笑话"}
+{"prompt": "Please generate a paraphrase for the following sentence.\ntwo boys reading superhero books ⇔ ", "response": "Two boys are reading books.", "predicted": -100, "true_label": "Two boys reading a piece of literature."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo boys reading a piece of literature. ⇔ ", "response": "Two boys with books.", "predicted": -100, "true_label": "two boys reading superhero books"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDark-haired man wearing a watch and oven mitt about to cook some meat in the kitchen. ⇔ ", "response": "A man is about to put meat in the oven.", "predicted": -100, "true_label": "A man is cooking something to eat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is cooking something to eat. ⇔ ", "response": "A man in an apron smiles as he cooks.", "predicted": -100, "true_label": "Dark-haired man wearing a watch and oven mitt about to cook some meat in the kitchen."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOa8是4G网吗 ⇔ ", "response": "oppoa8支持4g吗", "predicted": 8, "true_label": "gameboost"}
+{"prompt": "Please generate a paraphrase for the following sentence.\ngameboost ⇔ ", "response": "ZMA", "predicted": -100, "true_label": "OPPOa8是4G网吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n情人能做一辈子吗 ⇔ ", "response": "情人能不能长久？", "predicted": -100, "true_label": "情人能做一辈子吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n情人能做一辈子吗？ ⇔ ", "response": "情人之间能长久吗？", "predicted": -100, "true_label": "情人能做一辈子吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n1OPPOWatch41mm ⇔ ", "response": "oppowatch41mm", "predicted": 41, "true_label": "k1r15xbom"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nk1r15xbom ⇔ ", "response": "kr15xbom", "predicted": 15, "true_label": "1OPPOWatch41mm"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n成都的服务中心在哪？ ⇔ ", "response": "深圳的服务中心和配件点在哪里？", "predicted": -100, "true_label": "客服服务中心在哪？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n客服服务中心在哪？ ⇔ ", "response": "客服服务中心", "predicted": -100, "true_label": "成都的服务中心在哪？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her. ⇔ ", "response": "A girl is smiling.", "predicted": -100, "true_label": "The girl is smiling because she is having fun."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe girl is smiling because she is having fun. ⇔ ", "response": "A girl in a yellow dress smiles as she jumps in the dirt between two large rocks.", "predicted": -100, "true_label": "A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n凌晨七点叫我起床 ⇔ ", "response": "早上七点叫我起床", "predicted": -100, "true_label": "帮我定个五点半的闹钟"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n帮我定个五点半的闹钟 ⇔ ", "response": "小布明天早上帮我闹钟", "predicted": -100, "true_label": "凌晨七点叫我起床"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n猫从一盘水中取出一个物体。 ⇔ ", "response": "一个猫从水中取出一个球并抓起一个球。", "predicted": -100, "true_label": "那只猫想吃玉米棒上的玉米。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n那只猫想吃玉米棒上的玉米。 ⇔ ", "response": "那只猫想要一根玉米。", "predicted": -100, "true_label": "猫从一盘水中取出一个物体。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSTG官员直接支持围攻等事件中的专业战术警察，包括作战，谈判，情报和指挥支援服务。 ⇔ ", "response": "STG官员直接支持专业战术警察，包括作战，谈判，情报和指挥支援服务，事件如围攻。", "predicted": -100, "true_label": "STG官员在围攻等事件中支持战术警察，直接提供作战，谈判，情报和指挥支援服务。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSTG官员在围攻等事件中支持战术警察，直接提供作战，谈判，情报和指挥支援服务。 ⇔ ", "response": "STG官员直接提供作战，谈判，情报和指挥支援服务，在围攻等事件中支持战术警察。", "predicted": -100, "true_label": "STG官员直接支持围攻等事件中的专业战术警察，包括作战，谈判，情报和指挥支援服务。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃草菇吗 ⇔ ", "response": "怀孕能吃蘑菇吗", "predicted": -100, "true_label": "怀孕能吃鸡翅吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃鸡翅吗 ⇔ ", "response": "怀孕能吃鸡油菌吗", "predicted": -100, "true_label": "怀孕能吃草菇吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n曼德拉被家人缅怀，因为体育场的追悼会正在筹划中。 ⇔ ", "response": "曼德拉逝世后，他的家人正在南非各地受到缅怀。", "predicted": -100, "true_label": "曼德拉的家人感谢国家和世界的支持"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n曼德拉的家人感谢国家和世界的支持 ⇔ ", "response": "祖鲁说，曼德拉的家人感谢国家的支持", "predicted": -100, "true_label": "曼德拉被家人缅怀，因为体育场的追悼会正在筹划中。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n蝙蝠侠和罗宾驾驶直升机飞过水面。 ⇔ ", "response": "蝙蝠侠和罗宾在水面上飞行。", "predicted": -100, "true_label": "一架直升飞机在水面上空飞行。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一架直升飞机在水面上空飞行。 ⇔ ", "response": "一架直升机在水面上空飞行。", "predicted": -100, "true_label": "蝙蝠侠和罗宾驾驶直升机飞过水面。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n是的，是可以提前还款也可以自动还款的。 ⇔ ", "response": "是的，是可以自动还款 also: 也可以提前还款的。", "predicted": -100, "true_label": "您好，可以自动还款也可以主动还款的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n您好，可以自动还款也可以主动还款的 ⇔ ", "response": "您的信用卡可用来还款其他信用卡吗", "predicted": -100, "true_label": "是的，是可以提前还款也可以自动还款的。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你喜欢张杰吗 ⇔ ", "response": "喜欢张杰", "predicted": -100, "true_label": "王者荣耀蔡文姬"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n王者荣耀蔡文姬 ⇔ ", "response": "王者荣耀蔡文姬", "predicted": -100, "true_label": "你喜欢张杰吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe girl is using an eye pencil on her eyelid ⇔ ", "response": "A girl is doodling with an eye pencil on her eyelid", "predicted": -100, "true_label": "The girl is not using an eye pencil on her eyelid"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe girl is not using an eye pencil on her eyelid ⇔ ", "response": "The girl is using an eye pencil on her eyelid", "predicted": -100, "true_label": "The girl is using an eye pencil on her eyelid"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThis stimulates the interest of Russian oil companies to the construction of an oil pipeline from Western Siberia to the Pacific Ocean. ⇔ ", "response": "An oil pipeline is constructed in Western Siberia.", "predicted": -100, "true_label": "An oil pipeline in Siberia was ruptured."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn oil pipeline in Siberia was ruptured. ⇔ ", "response": "On September 10, 700 tonnes of oil spilled into the Vatyaun and Pim rivers in Russia's Western Siberia due to a local pipeline rupture.", "predicted": 10, "true_label": "This stimulates the interest of Russian oil companies to the construction of an oil pipeline from Western Siberia to the Pacific Ocean."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n美国时间现在几点 ⇔ ", "response": "现在什么时间，美国", "predicted": -100, "true_label": "除夕是什么时候"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n除夕是什么时候 ⇔ ", "response": "你有除夕吗", "predicted": -100, "true_label": "美国时间现在几点"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe lady is cracking an egg for the mixer ⇔ ", "response": "The lady is cracking two eggs into a mixer", "predicted": -100, "true_label": "The lady isn 't cracking an egg for the mixer"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe lady isn 't cracking an egg for the mixer ⇔ ", "response": "The lady is cracking an egg for the mixer", "predicted": -100, "true_label": "The lady is cracking an egg for the mixer"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe lady is slicing up the meat ⇔ ", "response": "The lady is carefully slicing up the meat", "predicted": -100, "true_label": "The lady isn 't slicing up the meat"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe lady isn 't slicing up the meat ⇔ ", "response": "The lady is slicing up the meat", "predicted": -100, "true_label": "The lady is slicing up the meat"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n韭菜有什么好处 ⇔ ", "response": "韭菜好不好吃", "predicted": -100, "true_label": "多吃韭菜有什么好处"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n多吃韭菜有什么好处 ⇔ ", "response": "多吃韭菜有什么坏处", "predicted": -100, "true_label": "韭菜有什么好处"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今日步数 ⇔ ", "response": "今天步数", "predicted": -100, "true_label": "帮我刷步数"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n帮我刷步数 ⇔ ", "response": "怎么刷步数", "predicted": -100, "true_label": "今日步数"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA dog is playing with a toy ⇔ ", "response": "A dog is not playing with a toy", "predicted": -100, "true_label": "A dog is having fun with a toy"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA dog is having fun with a toy ⇔ ", "response": "There is no dog playing with a ball", "predicted": -100, "true_label": "A dog is playing with a toy"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA soccer player is scoring a goal ⇔ ", "response": "A soccer player is not scoring a goal", "predicted": -100, "true_label": "There are no soccer players kicking a soccer ball into the goal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are no soccer players kicking a soccer ball into the goal ⇔ ", "response": "A soccer player is kicking a soccer ball into the goal", "predicted": -100, "true_label": "A soccer player is scoring a goal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nhifi ⇔ ", "response": "hich", "predicted": -100, "true_label": "hihi"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nhihi ⇔ ", "response": "hello", "predicted": -100, "true_label": "hifi"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n《千金女贼》中的片尾曲叫什么 ⇔ ", "response": "《女贼》片尾曲叫什么", "predicted": -100, "true_label": "千金女贼的片尾曲是什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n千金女贼的片尾曲是什么 ⇔ ", "response": "女贼的千金结局是什么", "predicted": -100, "true_label": "《千金女贼》中的片尾曲叫什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有哪些名言 ⇔ ", "response": "讲什么意义", "predicted": -100, "true_label": "适合大哥的经典语录"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n适合大哥的经典语录 ⇔ ", "response": "适合你妹的经典语录", "predicted": -100, "true_label": "有哪些名言"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn earthquake has hit the east coast of Hokkaido, Japan, with a magnitude of 7.0 Mw. ⇔ ", "response": "An earthquake has hit the east coast of Japan.", "predicted": -100, "true_label": "An earthquake occurred on the east coast of Hokkaido."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn earthquake occurred on the east coast of Hokkaido. ⇔ ", "response": "Startled East Coast residents from North Carolina to Maryland were shaken Tuesday by an earthquake that registered a preliminary magnitude of 4.5.", "predicted": 4, "true_label": "An earthquake has hit the east coast of Hokkaido, Japan, with a magnitude of 7.0 Mw."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe monkey isn 't practicing martial arts ⇔ ", "response": "A monkey is practicing martial arts", "predicted": -100, "true_label": "The monkey is practicing martial arts"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe monkey is practicing martial arts ⇔ ", "response": "A monkey is doing a pushup", "predicted": -100, "true_label": "The monkey isn 't practicing martial arts"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个男人捡起一块大石头。 ⇔ ", "response": "一个男人正在移动石头。", "predicted": -100, "true_label": "一个女人捡起一块大石头。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人捡起一块大石头。 ⇔ ", "response": "一个男人在切一些蔬菜。", "predicted": -100, "true_label": "一个男人捡起一块大石头。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公园散步 ⇔ ", "response": "昨天晚上我在家看电影", "predicted": -100, "true_label": "我公圆散布今天去"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我公圆散布今天去 ⇔ ", "response": "我公圆散布今天来", "predicted": -100, "true_label": "我今天去公园散步"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的世界什么都有的秘籍 ⇔ ", "response": "我的世界有什么秘籍", "predicted": -100, "true_label": "我的世界秘籍大全，手机版"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的世界秘籍大全，手机版 ⇔ ", "response": "我的世界手机版秘籍", "predicted": -100, "true_label": "我的世界什么都有的秘籍"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIf this can be confirmed, it may be possible to develop substances that could combat schizophrenia by neutralizing the schizophrenia antibodies. ⇔ ", "response": "Schizophrenia antibodies cause damage to cells.", "predicted": -100, "true_label": "Drugs are used to combat schizophrenia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDrugs are used to combat schizophrenia. ⇔ ", "response": "Schizophrenia is a severe psychiatric illness that affects 1% of people worldwide. Drugs used to combat the condition are less than satisfactory: developed through trial and error, many patients react badly to them and there is a growing body of evidence that they have a range of side-effects, including diabetes.", "predicted": 1, "true_label": "If this can be confirmed, it may be possible to develop substances that could combat schizophrenia by neutralizing the schizophrenia antibodies."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe boys are playing with Legos. ⇔ ", "response": "The children are a group of boys sitting down and building something out of Legos.", "predicted": -100, "true_label": "Boys are playing."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBoys are playing. ⇔ ", "response": "Boys wearing kimonos are playing sumo wrestling.", "predicted": -100, "true_label": "The boys are playing with Legos."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问印章里的字是什么？ ⇔ ", "response": "请问是什么？", "predicted": -100, "true_label": "请问这个印章刻的是什么字？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这个印章刻的是什么字？ ⇔ ", "response": "请问这是哪个国家的印章？", "predicted": -100, "true_label": "请问印章里的字是什么？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl in a black sweater and jeans pours water into an empty soda bottle. ⇔ ", "response": "A girl is filling a soda bottle.", "predicted": -100, "true_label": "A girl pours water into an empty coca cola bottle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl pours water into an empty coca cola bottle ⇔ ", "response": "A girl is filling up a bottle", "predicted": -100, "true_label": "A girl in a black sweater and jeans pours water into an empty soda bottle."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person in a large pink triangle-shaped costume waves as a man and a woman dressed in matching brightly-colored striped and polka-dotted outfits smile back. ⇔ ", "response": "A person in a costume is waving.", "predicted": -100, "true_label": "The person is moving her hand."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe person is moving her hand. ⇔ ", "response": "A woman with a green headscarf and no arms is moving her hand.", "predicted": -100, "true_label": "A person in a large pink triangle-shaped costume waves as a man and a woman dressed in matching brightly-colored striped and polka-dotted outfits smile back."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhile they did not find any association between cancer rates and consumption of caffeinated coffee or tea, people who regularly drank two or more cups per day of decaffeinated coffee had about half the incidence of rectal cancer as those who never drank decaf. ⇔ ", "response": "Decaffeinated coffee is associated with a reduced incidence of rectal cancer.", "predicted": -100, "true_label": "Tea protects from some diseases."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTea protects from some diseases. ⇔ ", "response": "Scientists have discovered that drinking tea protects against heart disease by improving the function of the artery walls.", "predicted": -100, "true_label": "While they did not find any association between cancer rates and consumption of caffeinated coffee or tea, people who regularly drank two or more cups per day of decaffeinated coffee had about half the incidence of rectal cancer as those who never drank decaf."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA pro-baseball pitcher sends a curve ball to his opponent while many fans relax and enjoy the game. ⇔ ", "response": "A pitcher is playing baseball at a stadium", "predicted": -100, "true_label": "A pitcher is playing baseball"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA pitcher is playing baseball ⇔ ", "response": "A man is playing baseball", "predicted": -100, "true_label": "A pro-baseball pitcher sends a curve ball to his opponent while many fans relax and enjoy the game."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这个是一一个测试案例 ⇔ ", "response": "这个是一一个测试", "predicted": -100, "true_label": "这是一个测试案例"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这是一个测试案例 ⇔ ", "response": "这不是一个案例", "predicted": -100, "true_label": "这个是一一个测试案例"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" I have lots of bad dreams , I have flashbacks , I have lots of anger . ⇔ ", "response": "\"I have lots of flashbacks , I have anger , I have bad dreams .", "predicted": -100, "true_label": "\" I have lots of bad dreams , flashbacks and lots of anger . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" I have lots of bad dreams , flashbacks and lots of anger . \" ⇔ ", "response": "You ever feel like crying ? I feel like crying a lot .", "predicted": -100, "true_label": "\" I have lots of bad dreams , I have flashbacks , I have lots of anger ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA red truck is parked next to a burning blue building while a man in a green vest runs toward it. ⇔ ", "response": "A man is running towards a burning building from a red truck.", "predicted": -100, "true_label": "The burning blue building smells of smoke"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe burning blue building smells of smoke ⇔ ", "response": "The smoldering blue building smells of smoke", "predicted": -100, "true_label": "A red truck is parked next to a burning blue building while a man in a green vest runs toward it."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA guy in a blue shirt chases a guy in an orange shirt during a game of flag football. ⇔ ", "response": "Two people are playing flag football.", "predicted": -100, "true_label": "Two people are playing flag football at the park."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people are playing flag football at the park. ⇔ ", "response": "Two people are playing flags at the park.", "predicted": -100, "true_label": "A guy in a blue shirt chases a guy in an orange shirt during a game of flag football."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nABCDEFJ ⇔ ", "response": "Abcdef", "predicted": -100, "true_label": "HiJerry"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHiJerry ⇔ ", "response": "hi", "predicted": -100, "true_label": "ABCDEFJ"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n布莱克本于2017年7月1日被召入亚特兰大勇士队，在大联盟对阵奥克兰田径队的比赛中首次亮相。 ⇔ ", "response": "布莱克本于2017年7月1日被召入亚特兰大勇士队，在大联盟对阵奥克兰田径队的比赛中首次亮相。", "predicted": 2017, "true_label": "布莱克本被邀请到亚特兰大勇士队，在2017年7月1日对阵奥克兰运动家队时首次亮相。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n布莱克本被邀请到亚特兰大勇士队，在2017年7月1日对阵奥克兰运动家队时首次亮相。 ⇔ ", "response": "布莱克本被邀请到亚特兰大勇士队，在2017年7月1日与奥克兰运动家队的比赛中首次亮相。", "predicted": 2017, "true_label": "布莱克本于2017年7月1日被召入亚特兰大勇士队，在大联盟对阵奥克兰田径队的比赛中首次亮相。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is brushing a cat ⇔ ", "response": "A person is brushing a hair off a cat's head", "predicted": -100, "true_label": "Nobody is brushing a cat"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNobody is brushing a cat ⇔ ", "response": "A woman is brushing a cat", "predicted": -100, "true_label": "A person is brushing a cat"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMH133 ⇔ ", "response": "M31", "predicted": 31, "true_label": "mh135"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nmh135 ⇔ ", "response": "h135m", "predicted": 135, "true_label": "MH133"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天的天气看起来非常好 ⇔ ", "response": "今天的天气看起来非常好", "predicted": -100, "true_label": "明天的天气看起来飞常好"}

eval_gen/gpu_results/results_gpu_1.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天的天气看起来飞常好 ⇔ ", "response": "我喜欢阅读科幻小说", "predicted": -100, "true_label": "明天的天气看起来非常好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is slicing an eggplant ⇔ ", "response": "There is no woman slicing an eggplant", "predicted": -100, "true_label": "A woman is slicing a vegetable"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is slicing a vegetable ⇔ ", "response": "There is no woman slicing a vegetable", "predicted": -100, "true_label": "A woman is slicing an eggplant"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么更改来电提示音？ ⇔ ", "response": "如何改变默认的来电音", "predicted": -100, "true_label": "如何更改充电提示音？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n如何更改充电提示音？ ⇔ ", "response": "充电提示音如何更改？", "predicted": -100, "true_label": "怎么更改来电提示音？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe number of extremely obese adults -- those who are at least 100 pounds overweight -- has quadrupled since the 1980s to about 4 million . ⇔ ", "response": "In 1985, there were 2.2 million very obese adults in the United States -- those who were at least 100 pounds above a healthy weight . Now , that number has quadrupled .", "predicted": 1985, "true_label": "The number of Americans considered extremely obese , or at least 100 pounds overweight , has quadrupled since the 1980s to a startling 4 million , the research shows ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe number of Americans considered extremely obese , or at least 100 pounds overweight , has quadrupled since the 1980s to a startling 4 million , the research shows . ⇔ ", "response": "Researchers say the study is the first to use the new criteria to estimate the number of Americans who are extremely obese . The criteria were developed by the National Institutes of Health and the American Association for the Study of Obesity . They define extremely obese people as those who are at least 100 pounds overweight and measure at least 49 inches ( 1.25 meters ) in height or have a waist size of 40 inches ( 1.02 meters ) or more . The researchers used data from the National Health and Nutrition Examination Survey to estimate the number of Americans who meet those criteria .", "predicted": 100, "true_label": "The number of extremely obese adults -- those who are at least 100 pounds overweight -- has quadrupled since the 1980s to about 4 million ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in glasses and an orange shirt is sitting outside next to a table with drinks on it. ⇔ ", "response": "A man is sitting outside.", "predicted": -100, "true_label": "A man sits near a table with his drinks"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man sits near a table with his drinks ⇔ ", "response": "A man , who is holding his drink , is sitting on the couch .", "predicted": -100, "true_label": "A man in glasses and an orange shirt is sitting outside next to a table with drinks on it."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n如何让自己的心变得强大呢 ⇔ ", "response": "如何让自己心情好？", "predicted": -100, "true_label": "怎么让自己变得强大"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么让自己变得强大 ⇔ ", "response": "怎么让自己强", "predicted": -100, "true_label": "如何让自己的心变得强大呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他昨天去商店买了一些苹果的。 ⇔ ", "response": "他昨天去商店买了一些苹果的，一篮子的。", "predicted": -100, "true_label": "他昨天去商店买了一些苹。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他昨天去商店买了一些苹。 ⇔ ", "response": "他今天去商店买了一些苹果。", "predicted": -100, "true_label": "他昨天去商店买了一些苹果的。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is not cooking something ⇔ ", "response": "The woman is cooking something", "predicted": -100, "true_label": "A woman is cooking eggs"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is cooking eggs ⇔ ", "response": "The woman is cooking something", "predicted": -100, "true_label": "The woman is not cooking something"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA monkey is brushing the dog ⇔ ", "response": "The dog is being brushed by a monkey", "predicted": -100, "true_label": "The monkey is not brushing a bull dog"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe monkey is not brushing a bull dog ⇔ ", "response": "The monkey is brushing a bull dog", "predicted": -100, "true_label": "A monkey is brushing the dog"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe doctors are reviving a man ⇔ ", "response": "The doctors are not reviving a man", "predicted": -100, "true_label": "The doctor is helping the patient"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe doctor is helping the patient ⇔ ", "response": "The doctors are healing a man", "predicted": -100, "true_label": "The doctors are reviving a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone is scratching the belly of an animal ⇔ ", "response": "The woman is scratching the animal's belly", "predicted": -100, "true_label": "Someone is stroking the belly of an animal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone is stroking the belly of an animal ⇔ ", "response": "The woman is stroking the animal's belly", "predicted": -100, "true_label": "Someone is scratching the belly of an animal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么黑人喜欢吃白巧克力 ⇔ ", "response": "白巧克力是什么", "predicted": -100, "true_label": "黑人为什么喜欢吃白巧克力?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n黑人为什么喜欢吃白巧克力? ⇔ ", "response": "白巧克力是谁发明的", "predicted": -100, "true_label": "为什么黑人喜欢吃白巧克力"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo boy are playing on a trampoline ⇔ ", "response": "A boy is jumping on a trampoline", "predicted": -100, "true_label": "Nobody is playing on a trampoline"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNobody is playing on a trampoline ⇔ ", "response": "A trampoline is being played with by a kid", "predicted": -100, "true_label": "Two boy are playing on a trampoline"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他去商店买了一些苹果和橙子，然而并没有去银行取钱。 ⇔ ", "response": "他去商店买了一些苹果和橙子，然而并没有去银行。", "predicted": -100, "true_label": "他去商店买了一些苹果和橙子，然而并没有去把行取钱。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他去商店买了一些苹果和橙子，然而并没有去把行取钱。 ⇔ ", "response": "他去商店买了一些苹果和橙子，然而并没有去支付 money。", "predicted": -100, "true_label": "他去商店买了一些苹果和橙子，然而并没有去银行取钱。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n哪里有手工包批发啊？ ⇔ ", "response": "哪里有手工包批发", "predicted": -100, "true_label": "哪里有手工包批发？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n哪里有手工包批发？ ⇔ ", "response": "哪里有手工包批发", "predicted": -100, "true_label": "哪里有手工包批发啊？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n核不扩散条约已有37年的历史。 ⇔ ", "response": "核不扩散条约自1968年以来一直存在。", "predicted": 1968, "true_label": "核不扩散条约每5年审查一次。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n核不扩散条约每5年审查一次。 ⇔ ", "response": "核不扩散条约三年审查一次。", "predicted": -100, "true_label": "核不扩散条约已有37年的历史。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFIRSTDAY ⇔ ", "response": "Cutie", "predicted": -100, "true_label": "Onlyday"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOnlyday ⇔ ", "response": "NOWday", "predicted": -100, "true_label": "FIRSTDAY"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is playing a bamboo flute ⇔ ", "response": "A flute is being played by the man", "predicted": -100, "true_label": "The man is not playing a flute"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is not playing a flute ⇔ ", "response": "A man is playing a flute", "predicted": -100, "true_label": "A man is playing a bamboo flute"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n支付宝支付密码忘记了怎么办啊 ⇔ ", "response": "支付宝支付密码怎么记不得了", "predicted": -100, "true_label": "支付宝支付密码忘记了怎么办"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n支付宝支付密码忘记了怎么办 ⇔ ", "response": "支付宝支付密码怎么记回来了", "predicted": -100, "true_label": "支付宝支付密码忘记了怎么办啊"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他问。以前我回答过了，因为你不爱我。 ⇔ ", "response": "他问。因为我爱你，所以我不回答。", "predicted": -100, "true_label": "我回答，是因为你不爱我。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我回答，是因为你不爱我。 ⇔ ", "response": "我这样回答，是因为你不喜欢我。", "predicted": -100, "true_label": "他问。以前我回答过了，因为你不爱我。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n搜索一篇作文 ⇔ ", "response": "搜索一篇文章", "predicted": -100, "true_label": "植树节是通过什么形式确定的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n植树节是通过什么形式确定的 ⇔ ", "response": "中国植树节是通过什么形式确定的？", "predicted": -100, "true_label": "搜索一篇作文"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nForecasters predict the storm will reach Earth at 3 p.m. eastern time Friday and could last up to 18 hours . ⇔ ", "response": "The storm could hit Earth at 3 p.m. on Friday and could last up to 18 hours .", "predicted": 3, "true_label": "It is expected to reach Earth about 3 p.m. EDT Friday , and its effects could last 12 to 18 hours ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIt is expected to reach Earth about 3 p.m. EDT Friday , and its effects could last 12 to 18 hours . ⇔ ", "response": "The asteroid will pass about 26,000 miles above Earth 's surface at 1 p.m. EDT Friday , and its effects could last 12 to 18 hours .", "predicted": 26, "true_label": "Forecasters predict the storm will reach Earth at 3 p.m. eastern time Friday and could last up to 18 hours ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她紧紧地抱着多罗脱奥说:你为什么要把我弄醒，我的好姐姐啊？ ⇔ ", "response": "她紧紧地抱着多罗脱奥说:你为什么要把我弄醒，我的好姐姐啊？", "predicted": -100, "true_label": "她紧紧地抱住多罗特亚，说:我可爱的夫人呀，你为什么要把我叫醒呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她紧紧地抱住多罗特亚，说:我可爱的夫人呀，你为什么要把我叫醒呢？ ⇔ ", "response": "她说:我可爱的夫人呀，你为什么要把我叫醒呢？", "predicted": -100, "true_label": "她紧紧地抱着多罗脱奥说:你为什么要把我弄醒，我的好姐姐啊？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第28步兵师的第2个BCT集中在拉马迪控制主要道路并保护州长和政府中心。 ⇔ ", "response": "第28步兵师的第2个BCT集中在拉马迪控制主要道路和保护州长和政府中心。", "predicted": 28, "true_label": "第28步兵师的第二个BCT专注于拉马迪，保护主要道路，控制州长和政府中心。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第28步兵师的第二个BCT专注于拉马迪，保护主要道路，控制州长和政府中心。 ⇔ ", "response": "第28步兵师的第二个BCT专注于拉马迪，保护主要道路，控制州长和政府中心。", "predicted": 28, "true_label": "第28步兵师的第2个BCT集中在拉马迪控制主要道路并保护州长和政府中心。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA golfer is getting ready to putt on the green, with a crowd of people watching in the background. ⇔ ", "response": "A golfer is putting on a green, with a crowd of people watching.", "predicted": -100, "true_label": "The golfer is getting ready to putt on the green, with a crowd watching in the background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe golfer is getting ready to putt on the green, with a crowd watching in the background. ⇔ ", "response": "A crowd is watching a golfer putt.", "predicted": -100, "true_label": "A golfer is getting ready to putt on the green, with a crowd of people watching in the background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man eating some food ⇔ ", "response": "A man is eating some food", "predicted": -100, "true_label": "A man is eating some food"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is eating some food ⇔ ", "response": "There is no man eating food", "predicted": -100, "true_label": "There is no man eating some food"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n头发怎么变软呢?自己在家就能操作的 ⇔ ", "response": "怎么让头发自然卷？", "predicted": -100, "true_label": "我的头发偏硬，怎么采能变软"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的头发偏硬，怎么采能变软 ⇔ ", "response": "为什么我的头发越吹越硬", "predicted": -100, "true_label": "头发怎么变软呢?自己在家就能操作的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你脑子进水了吗 ⇔ ", "response": "你有脑子吗", "predicted": -100, "true_label": "���吧我爱你"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n好吧我爱你 ⇔ ", "response": "你什么？爱你", "predicted": -100, "true_label": "你脑子进水了吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在他的妻子去世后，约翰比阿特丽斯与波旁人结婚，他们有两个孩子，卢森堡公爵瓦茨拉夫一世和年轻时去世的波恩。 ⇔ ", "response": "约翰比阿特丽斯与波旁人结婚，他们有两个孩子，卢森堡公爵瓦茨拉夫一世和早逝的波恩。", "predicted": -100, "true_label": "妻子去世后，约翰与波旁的比阿特丽斯结婚。他们有两个孩子，卢森堡公爵Wenceslaus I和年轻时去世的Bonne。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n妻子去世后，约翰与波旁的比阿特丽斯结婚。他们有两个孩子，卢森堡公爵Wenceslaus I和年轻时去世的Bonne。 ⇔ ", "response": "妻子去世后，约翰与波旁的比阿特丽斯结婚，他们有两个孩子，卢森堡公爵Wenceslaus I和早逝的Bonne。", "predicted": -100, "true_label": "在他的妻子去世后，约翰比阿特丽斯与波旁人结婚，他们有两个孩子，卢森堡公爵瓦茨拉夫一世和年轻时去世的波恩。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn old man looking over a sculpture. ⇔ ", "response": "A man in a hat looking at a statue.", "predicted": -100, "true_label": "the man is old"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nthe man is old ⇔ ", "response": "An old man is sitting in a green chair.", "predicted": -100, "true_label": "An old man looking over a sculpture."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThose conversations had not taken place as of Tuesday night , according to an Oracle spokeswoman . ⇔ ", "response": "\" Those conversations had not taken place as of Tuesday night , \" toledo.com , an Oracle affiliate , reported , to which an Oracle spokeswoman declined to comment .", "predicted": -100, "true_label": "Those talks have not taken place , according to an Oracle spokeswoman ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThose talks have not taken place , according to an Oracle spokeswoman . ⇔ ", "response": "But no such talks have taken place , according to an Oracle spokeswoman .", "predicted": -100, "true_label": "Those conversations had not taken place as of Tuesday night , according to an Oracle spokeswoman ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFederal prosecutors , the Securities and Exchange Commission , and the mortgage company 's regulator launched investigations into Freddie Mac 's shake-up . ⇔ ", "response": "Federal prosecutors , the Securities and Exchange Commission and the mortgage company 's regulator launched investigations into the shake-up .", "predicted": -100, "true_label": "The Securities and Exchange Commission and the U.S. Attorney have opened investigations into Freddie Mac over its accounting practices ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe Securities and Exchange Commission and the U.S. Attorney have opened investigations into Freddie Mac over its accounting practices . ⇔ ", "response": "The SEC and federal prosecutors have started investigating Freddie Mac 's ( Federal Home Loan Mortgage Corp. ) accounting practices .", "predicted": -100, "true_label": "Federal prosecutors , the Securities and Exchange Commission , and the mortgage company 's regulator launched investigations into Freddie Mac 's shake-up ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn a stunning setback for Bill Gates' software empire, the judge in the Microsoft antitrust trial ruled yesterday that the software giant is a monopoly that wielded its power to stifle competition. ⇔ ", "response": "Bill Gates controls Microsoft.", "predicted": -100, "true_label": "Microsoft holds a monopoly power."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMicrosoft holds a monopoly power. ⇔ ", "response": "Microsoft maintains their very successful operating system monopoly through the use of proprietary specifications as well as aggressive marketing.", "predicted": -100, "true_label": "In a stunning setback for Bill Gates' software empire, the judge in the Microsoft antitrust trial ruled yesterday that the software giant is a monopoly that wielded its power to stifle competition."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThat department, a co-manager of shipwrecks with the other two, has also attempted to address its shipwreck responsibilities since the early '70s. ⇔ ", "response": "A department co-managers shipwrecks.", "predicted": -100, "true_label": "Shipwreck salvaging was attempted."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nShipwreck salvaging was attempted. ⇔ ", "response": "Hundreds of divers and treasure hunters, including the Duke of Argyll, have risked their lives in the dangerous waters of the Isle of Mull trying to discover the reputed 30,000,000 pounds in Gold carried by this vessel--the target of the most enduring treasure hunt in British history.", "predicted": 30, "true_label": "That department, a co-manager of shipwrecks with the other two, has also attempted to address its shipwreck responsibilities since the early '70s."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公园散布了。 ⇔ ", "response": "我今天去公园散步了。", "predicted": -100, "true_label": "我今天去公园散步聊。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公园散步聊。 ⇔ ", "response": "我今天去公园散步。", "predicted": -100, "true_label": "我今天去公园散布了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\ncake ⇔ ", "response": "cake1", "predicted": 1, "true_label": "qcakecakecakecake"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nqcakecakecakecake ⇔ ", "response": "cake", "predicted": -100, "true_label": "cake"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRetailers J.C. Penney Co . Inc.JCP.N and Walgreen Co.WAG.N kick things off early in the week . ⇔ ", "response": "General Retailers J.C. Penney Co . Inc.JCP.N and Walgreen Co.WAG.N are setting July 4th as the launch date for their annual sales .", "predicted": 4, "true_label": "Retailers J.C. Penney Co . Inc . JCP.N and Walgreen Co . WAG.N kick things off on Monday ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRetailers J.C. Penney Co . Inc . JCP.N and Walgreen Co . WAG.N kick things off on Monday . ⇔ ", "response": "The bids conclude Monday , with results Tuesday .", "predicted": -100, "true_label": "Retailers J.C. Penney Co . Inc.JCP.N and Walgreen Co.WAG.N kick things off early in the week ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么查别人微信记录 ⇔ ", "response": "别人微信记录我怎么查", "predicted": -100, "true_label": "怎么查别人微信记录？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么查别人微信记录？ ⇔ ", "response": "别人手机微信记录怎么查？", "predicted": -100, "true_label": "怎么查别人微信记录"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个男人正系在速记机上。 ⇔ ", "response": "一个男人在记什么。", "predicted": -100, "true_label": "那人用速记。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n那人用速记。 ⇔ ", "response": "那人正在画速记。", "predicted": -100, "true_label": "一个男人正系在速记机上。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他们的男朋友很有钱吗 ⇔ ", "response": "他们的男朋友什么样子", "predicted": -100, "true_label": "你很有钱吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你很有钱吗 ⇔ ", "response": "你有什么值得学习的吗", "predicted": -100, "true_label": "他们的男朋友很有钱吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天的上证指数 ⇔ ", "response": "今天的上证指数吧", "predicted": -100, "true_label": "法拉利多少钱"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n法拉利多少钱 ⇔ ", "response": "myfarailifixstohere", "predicted": -100, "true_label": "今天的上证指数"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head. ⇔ ", "response": "A man and woman are in a field.", "predicted": -100, "true_label": "A couple are having their wedding picture taken."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA couple are having their wedding picture taken. ⇔ ", "response": "A bride and groom as well as the couple's friends are standing outside in the snow, with the groom holding the bride by her dress.", "predicted": -100, "true_label": "A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n安逸哟，是什么意思。 ⇔ ", "response": "啥叫安逸。", "predicted": -100, "true_label": "好安逸是什么意思"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n好安逸是什么意思 ⇔ ", "response": "好安逸是什么", "predicted": -100, "true_label": "安逸哟，是什么意思。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么？因为在您和她的关系问题没有明确解决以前，她不好作出选择从我这方面说，问题已经得到明确��解决。 ⇔ ", "response": "为什么是那样呢？因为在您和她的关系问题没有明确解决以前，她不好作出选择从我的方面说，问题已经得到明确的解决，您不再是我的竞争者了。", "predicted": -100, "true_label": "因为在您跟她的关系问题没有最后解决以前，她不能做出其他任何抉择。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n因为在您跟她的关系问题没有最后解决以前，她不能做出其他任何抉择。 ⇔ ", "response": "因为在你和她的关系问题得到解决前，她是不可能做出其他选择的。", "predicted": -100, "true_label": "为什么？因为在您和她的关系问题没有明确解决以前，她不好作出选择从我这方面说，问题已经得到明确的解决。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn addition , the Justice Department said that the FBI has conducted ' ' fewer than 10 ' ' investigations involving visits to mosques . ⇔ ", "response": "The Justice Department said Friday that the FBI has conducted \" fewer than 10 \" investigations involving visits to mosques .", "predicted": 10, "true_label": "In addition , \" fewer than 10 \" FBI offices have conducted investigations involving visits to Islamic mosques , the Justice Department said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn addition , \" fewer than 10 \" FBI offices have conducted investigations involving visits to Islamic mosques , the Justice Department said . ⇔ ", "response": "\" Fewer than 10 \" FBI offices have conducted such investigations , the department said in a legal brief last week .", "predicted": 10, "true_label": "In addition , the Justice Department said that the FBI has conducted ' ' fewer than 10 ' ' investigations involving visits to mosques ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDrinking four or five cups of green tea a day can help protect against cancer and heart disease, aid in weight loss, ease arthritis and chronic pain and even help prevent Alzheimer's disease. ⇔ ", "response": "Green tea is used for preventing Alzheimer's disease.", "predicted": -100, "true_label": "Tea protects from some diseases."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTea protects from some diseases. ⇔ ", "response": "Scientists have discovered that drinking tea protects against heart disease by improving the function of the artery walls.", "predicted": -100, "true_label": "Drinking four or five cups of green tea a day can help protect against cancer and heart disease, aid in weight loss, ease arthritis and chronic pain and even help prevent Alzheimer's disease."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is pouring oil into a skillet ⇔ ", "response": "A man is pouring oil into a pan", "predicted": -100, "true_label": "A man is taking the oil out off a skillet"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is taking the oil out off a skillet ⇔ ", "response": "A man is taking the oil from a skillet", "predicted": -100, "true_label": "A man is pouring oil into a skillet"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n奥迪A四 ⇔ ", "response": "奥迪A6L", "predicted": 6, "true_label": "宝马五系和奥迪A4L有什么区别"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n宝马五系和奥迪A4L有什么区别 ⇔ ", "response": "宝马5系和4系有什么区别？", "predicted": 5, "true_label": "奥迪A四"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman squeezing juice out of a lemon ⇔ ", "response": "A woman is squeezing juice out of a lemon", "predicted": -100, "true_label": "A woman is squeezing a lemon"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is squeezing a lemon ⇔ ", "response": "A woman is not squeezing a lemon", "predicted": -100, "true_label": "There is no woman squeezing juice out of a lemon"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n播放一首桥边姑娘 ⇔ ", "response": "音乐桥边姑娘", "predicted": -100, "true_label": "播放酷酷狗音乐"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n播放酷酷狗音乐 ⇔ ", "response": "播放今天天气", "predicted": -100, "true_label": "播放一首桥边姑娘"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOn August 15, IDFA submitted its comments regarding the U.S. Department of Agriculture's (USDA) latest draft of specifications (known as \"Dairy 6\") for government purchases of salted butter, cheddar cheese (blocks), American-style cheese (barrels) and nonfat dry milk. ⇔ ", "response": "On August 15, IDFA bought US dairy.", "predicted": 15, "true_label": "A government purchases food."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA government purchases food. ⇔ ", "response": "Restricted items are those items that are not acceptable for purchase using government appropriated funds. Some examples of items that the General Accounting Office (GAO) considers \"restricted\" include: food, live entertainment, and personal gifts including plaques.", "predicted": -100, "true_label": "On August 15, IDFA submitted its comments regarding the U.S. Department of Agriculture's (USDA) latest draft of specifications (known as \"Dairy 6\") for government purchases of salted butter, cheddar cheese (blocks), American-style cheese (barrels) and nonfat dry milk."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn old woman is shaking hands with a man ⇔ ", "response": "There is no old woman shaking hands with a man", "predicted": -100, "true_label": "A man and a woman are shaking hands"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are shaking hands ⇔ ", "response": "A man and a woman are standing near each other and shaking hands", "predicted": -100, "true_label": "An old woman is shaking hands with a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n缝的时候，他两手发抖，但是尽力克制住，缝上以后，他又把大衣穿上，。 ⇔ ", "response": "缝制的时候，他两手发抖，但是尽力克制住，缝上以后，他又把大衣穿上。", "predicted": -100, "true_label": "他缝的时候两手发抖，但他克制住了，他缝得很好，再穿上大衣的时候，从外边一点也看不出来。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他缝的时候两手发抖，但他克制住了，他缝得很好，再穿上大衣的时候，从外边一点也看不出来。 ⇔ ", "response": "那个人在缝什么呢？是件大衣的里子吧？", "predicted": -100, "true_label": "缝的时候，他两手发抖，但是尽力克制住，缝上以后，他又把大衣穿上，。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOf the over 400 different species of sharks found in the world's oceans, only about 30 have been reported to ever attack a human. ⇔ ", "response": "A shark attacked a human being.", "predicted": -100, "true_label": "A shark attacked a human being."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA shark attacked a human being. ⇔ ", "response": "A shark encounter with a human typically consists of a shark leisurely circling and/or slowly swimming past the subject without any aggressive behavior being exhibited.", "predicted": -100, "true_label": "Of the over 400 different species of sharks found in the world's oceans, only about 30 have been reported to ever attack a human."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe herd of caribous is crossing a road ⇔ ", "response": "The herd of caribous is crossing a road quickly", "predicted": -100, "true_label": "The herd of caribous is not crossing a road"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe herd of caribous is not crossing a road ⇔ ", "response": "The herd of caribous is crossing a road", "predicted": -100, "true_label": "The herd of caribous is crossing a road"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA study of more than 90,000 Japanese found that people who drank coffee daily or nearly every day had half the liver cancer risk of those who never drank coffee. The protective effect occurred in people who drank one to two cups a day and increased at three to four cups. ⇔ ", "response": "Coffee causes liver cancer", "predicted": -100, "true_label": "Tea protects from some diseases."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTea protects from some diseases. ⇔ ", "response": "Scientists have discovered that drinking tea protects against heart disease by improving the function of the artery walls.", "predicted": -100, "true_label": "A study of more than 90,000 Japanese found that people who drank coffee daily or nearly every day had half the liver cancer risk of those who never drank coffee. The protective effect occurred in people who drank one to two cups a day and increased at three to four cups."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打开相机设置 ⇔ ", "response": "相机黑屏", "predicted": -100, "true_label": "照相机打开相机"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n照相机打开相机 ⇔ ", "response": "打开照相机", "predicted": -100, "true_label": "打开相机设置"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA head-on collision between a Dhaka bound bus and a motorcycle left three people dead at Gangabardi area on the Faridpur-Khulna Highway on September 26. ⇔ ", "response": "A bus hit a motorcycle.", "predicted": -100, "true_label": "People were killed in a collision between ships."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople were killed in a collision between ships. ⇔ ", "response": "On December 6, as a result of a collision between two ships in Halifax harbour, a cargo of ammunition exploded and 1,630 people lost their lives in the ensuing holocaust, thousands more being injured.", "predicted": 6, "true_label": "A head-on collision between a Dhaka bound bus and a motorcycle left three people dead at Gangabardi area on the Faridpur-Khulna Highway on September 26."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天的天气真不错 ⇔ ", "response": "我喜欢在周末看电影", "predicted": -100, "true_label": "今天的天气真不策"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天的天气真不策 ⇔ ", "response": "今天的天气真好啊", "predicted": -100, "true_label": "今天的天气真不错"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个男人在玩扑克牌。 ⇔ ", "response": "一个男人在切胡椒粉。", "predicted": -100, "true_label": "一个男人在玩纸牌戏法。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个男人在玩纸牌戏法。 ⇔ ", "response": "一个男人在玩纸牌。", "predicted": -100, "true_label": "一个男人在玩扑克牌。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n各种标志的含义 ⇔ ", "response": "标志意味着什么", "predicted": -100, "true_label": "这些标志的含义"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这些标志的含义 ⇔ ", "response": "标志意味着什么", "predicted": -100, "true_label": "各种标志的含义"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nEven today, within the deepest recesses of our mind, lies a primordial fear that will not allow us to enter the sea without thinking about the possibility of being attacked by a shark. ⇔ ", "response": "Sharks attack boats.", "predicted": -100, "true_label": "A shark attacked a human being."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA shark attacked a human being. ⇔ ", "response": "A shark encounter with a human typically consists of a shark leisurely circling and/or slowly swimming past the subject without any aggressive behavior being exhibited.", "predicted": -100, "true_label": "Even today, within the deepest recesses of our mind, lies a primordial fear that will not allow us to enter the sea without thinking about the possibility of being attacked by a shark."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么中国校服那么难看？ ⇔ ", "response": "为什么校服那么好看", "predicted": -100, "true_label": "中国校服为什么那么难看？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国校服为什么那么难看？ ⇔ ", "response": "中国的校服为什么那么难看啊", "predicted": -100, "true_label": "为什么中国校服那么难看？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo men spar in martial arts gear. ⇔ ", "response": "Two men in martial arts uniforms training together.", "predicted": -100, "true_label": "There are multiple people present."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are multiple people present. ⇔ ", "response": "A group of footballs players is standing behind a coaching official.", "predicted": -100, "true_label": "Two men spar in martial arts gear."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNobody is riding a bike ⇔ ", "response": "A bike is being ridden by someone", "predicted": -100, "true_label": "Two people are riding a bike"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people are riding a bike ⇔ ", "response": "Two people are stopping on a motorcycle", "predicted": -100, "true_label": "Nobody is riding a bike"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气很好 ⇔ ", "response": "我今天天气很", "predicted": -100, "true_label": "我今天天天气很好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天天气很好 ⇔ ", "response": "我今天天气很好", "predicted": -100, "true_label": "我今天天气很好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n语文的学习方法 ⇔ ", "response": "学习方法和技巧", "predicted": -100, "true_label": "学习语文的方法"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n学习语文的方法 ⇔ ", "response": "学习方法有哪些", "predicted": -100, "true_label": "语文的学习方法"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n腰酸背痛，而且很饿怎么回事 ⇔ ", "response": "腰酸背痛，很饿，吃什么都可以", "predicted": -100, "true_label": "腰酸背痛怎么回事啊"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n腰酸背痛怎么回事啊 ⇔ ", "response": "背痛怎么回事", "predicted": -100, "true_label": "腰酸背痛，而且很饿怎么回事"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n“以马内利”是什么意思？ ⇔ ", "response": "“在以色列中”是什么意思？", "predicted": -100, "true_label": "以马内利是什么意思?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n以马内利是什么意思? ⇔ ", "response": "以马内利在什么情况下使用", "predicted": -100, "true_label": "“以马内利”是什么意思？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe military officer is barking at the recruits ⇔ ", "response": "A soldier is barking at the recruits", "predicted": -100, "true_label": "An officer is talking to the recruits"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn officer is talking to the recruits ⇔ ", "response": "A recruit is being reprimanded by an officer", "predicted": -100, "true_label": "The military officer is barking at the recruits"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n温州售后电话 ⇔ ", "response": "珠海松下售后电话", "predicted": -100, "true_label": "电话放入电脑传送东西没回应"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n电话放入电脑传送东西没回应 ⇔ ", "response": "电脑传送东西没回应", "predicted": -100, "true_label": "温州售后电话"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我妹妹是谁 ⇔ ", "response": "我妹妹是誰", "predicted": -100, "true_label": "你知道我的妹妹叫什么名字吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你知道我的妹妹叫什么名字吗 ⇔ ", "response": "你知道我的名字吗", "predicted": -100, "true_label": "我妹妹是谁"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is eating a pizza ⇔ ", "response": "There is no man eating food", "predicted": -100, "true_label": "A man is eating some food"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is eating some food ⇔ ", "response": "There is no man eating food", "predicted": -100, "true_label": "A man is eating a pizza"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree men are sitting outside on chairs with red seats. ⇔ ", "response": "Three friends are outside sitting on chairs with red seats.", "predicted": -100, "true_label": "Men are sitting outside."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMen are sitting outside. ⇔ ", "response": "Men sitting on a bench with a woman in a white car.", "predicted": -100, "true_label": "Three men are sitting outside on chairs with red seats."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n沙特阿拉伯的死罪包括强奸、谋杀、叛教、武装抢劫和贩毒。 ⇔ ", "response": "沙特阿拉伯对谋杀、强奸、叛国、强盗和贩毒等罪行的死刑仍在等待执行。", "predicted": -100, "true_label": "伊朗的死罪包括谋杀、强奸、武装抢劫、严重贩毒和通奸。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n伊朗的死罪包括谋杀、强奸、武装抢劫、严重贩毒和通奸。 ⇔ ", "response": "伊朗的死罪被定义为谋杀、强奸、武装抢劫、严重贩毒和与犯有这些罪行的人有 sexual relations。", "predicted": -100, "true_label": "沙特阿拉伯的死罪包括强奸、谋杀、叛教、武装抢劫和贩毒。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na30 ⇔ ", "response": "a30wbt", "predicted": 30, "true_label": "OPPOA33"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOA33 ⇔ ", "response": "oppoa83", "predicted": 83, "true_label": "a30"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样制作表格 ⇔ ", "response": "制作表格", "predicted": -100, "true_label": "怎样学表格制作"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样学表格制作 ⇔ ", "response": "表格制作怎么用", "predicted": -100, "true_label": "怎样制作表格"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天气很好，我们可以去散步。 ⇔ ", "response": "今天天气很好，适合去散步。", "predicted": -100, "true_label": "天气很好，我们可已去散步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天气很好，我们可已去散步。 ⇔ ", "response": "天气很好，我们 Said 去散步。", "predicted": -100, "true_label": "天气很好，我们可以去散步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气很好 ⇔ ", "response": "我今天天气很", "predicted": -100, "true_label": "我今天天气很好好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天���很好好 ⇔ ", "response": "我今天天气很号号好", "predicted": -100, "true_label": "我今天天气很好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man with no shirt on is performing with a baton. ⇔ ", "response": "A man with no shirt on dances.", "predicted": -100, "true_label": "A man is doing things with a baton."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is doing things with a baton. ⇔ ", "response": "A man performing with a marimba baton.", "predicted": -100, "true_label": "A man with no shirt on is performing with a baton."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDecember ⇔ ", "response": "January", "predicted": -100, "true_label": "February"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFebruary ⇔ ", "response": "January", "predicted": -100, "true_label": "December"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child in a ninja outfit does a jumping kick. ⇔ ", "response": "a boy in a ninja costume doing a jumping kick", "predicted": -100, "true_label": "a child does a jumping kick"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na child does a jumping kick ⇔ ", "response": "A child plays.", "predicted": -100, "true_label": "A child in a ninja outfit does a jumping kick."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n语音助手无法语音解锁怎么办 ⇔ ", "response": "为什么语音解锁不了", "predicted": -100, "true_label": "语音助手无法开启"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n语音助手无法开启 ⇔ ", "response": "请问能开启什么", "predicted": -100, "true_label": "语音助手无法语音解锁怎么办"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is driving a car ⇔ ", "response": "There is no man driving a car", "predicted": -100, "true_label": "A man is driving a vehicle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is driving a vehicle ⇔ ", "response": "There is no man driving a vehicle", "predicted": -100, "true_label": "A man is driving a car"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n从你的全世界路过 ⇔ ", "response": "路过你的全世界", "predicted": -100, "true_label": "从你的全世界里路过"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n从你的全世界里路过 ⇔ ", "response": "路过你的全世界", "predicted": -100, "true_label": "从你的全世界路过"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天是什么日子 ⇔ ", "response": "明天是什么数", "predicted": -100, "true_label": "第七次全国人口普查摸底工作时间节点为"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第七次全国人口普查摸底工作时间节点为 ⇔ ", "response": "第七次全国人口普查正式调查时间为", "predicted": -100, "true_label": "明天是什么日子"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n癌前病变明显是典型的组织，在显微检镜查中看起来异常，并且相比其在形态上正常的对应物，癌症更有可能出现在其中。 ⇔ ", "response": "癌前病变是一种在显微检查中看起来异常的组织，相比其癌症性匹配组织，其癌症更有可能存在于其中。", "predicted": -100, "true_label": "癌前病变明显是一种典型组织，在用显微镜检查时其会呈现异常，而且在此处发生癌变的可能性大于其形态正常的对应组织。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n癌前病变明显是一种典型组织，在用显微镜检查时其会呈现异常，而且在此处发生癌变的可能性大于其形态正常的对应组织。 ⇔ ", "response": "在显微镜下检查明显癌前病变是一种典型组织，其用在此处发生癌变的地方可能性大于正常形态的组织。", "predicted": -100, "true_label": "癌前病变明显是典型的组织，在显微检镜查中看起来异常，并且相比其在形态上正常的对应物，癌症更有可能出现在其中。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n李现生日快乐 ⇔ ", "response": "李现的电视剧什么时候上映", "predicted": -100, "true_label": "王者荣耀刘备"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n王者荣耀刘备 ⇔ ", "response": "王者荣耀刘备皮肤", "predicted": -100, "true_label": "李现生日快乐"}

eval_gen/gpu_results/results_gpu_2.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天去看了一场电影，电影非常有趣。 ⇔ ", "response": "我昨天去看了一场电影，电影非常有趣。", "predicted": -100, "true_label": "我昨天去看了一场电影，电影非常有去。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天去看了一场电影，电影非常有去。 ⇔ ", "response": "我昨天去看了一场电影，电影非常好。", "predicted": -100, "true_label": "我昨天去看了一场电影，电影非常有趣。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我去散步了。 ⇔ ", "response": "今天天气真好，我去散布了。", "predicted": -100, "true_label": "今天天气真好，我去散步了。去 散布。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我去散步了。去 散布。 ⇔ ", "response": "今天天气真好，我去散步了。去散布。", "predicted": -100, "true_label": "今天天气真好，我去散步了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜欢吃苹郭果 ⇔ ", "response": "今天的天气真好", "predicted": -100, "true_label": "苹果是我喜欢的水果"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n苹果是我喜欢的水果 ⇔ ", "response": "天气很好，我们去公园散步吧", "predicted": -100, "true_label": "我喜欢吃苹郭果"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nILOVEYOUCHINA ⇔ ", "response": "Touchid", "predicted": -100, "true_label": "MadeinChina"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMadeinChina ⇔ ", "response": "SNOW", "predicted": -100, "true_label": "ILOVEYOUCHINA"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃灯笼椒吗 ⇔ ", "response": "能吃灯笼椒吗", "predicted": -100, "true_label": "橘子有什么好处"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n橘子有什么好处 ⇔ ", "response": "橘子为什么那么便宜", "predicted": -100, "true_label": "怀孕能吃灯笼椒吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去超市买了些蔬菜。 ⇔ ", "response": "我今天去超市买了些蔬菜啊。", "predicted": -100, "true_label": "我今天天超市买了些蔬菜。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天超市买了些蔬菜。 ⇔ ", "response": "我今天去超市买了些蔬菜。", "predicted": -100, "true_label": "我今天去超市买了些蔬菜。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThese folk art traditions have been preserved for hundreds of years. ⇔ ", "response": "As a result, these folk art traditions have been preserved for hundreds of years.", "predicted": -100, "true_label": "Indigenous folk art is preserved."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIndigenous folk art is preserved. ⇔ ", "response": "Due to Beijing's long history as an agrarian society, a rich and vibrant cultural heritage has been preserved. One part of this culture is the indigenous folk art and traditional handicrafts.", "predicted": -100, "true_label": "These folk art traditions have been preserved for hundreds of years."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n李白写了多少诗 ⇔ ", "response": "李白创作了多少首诗", "predicted": -100, "true_label": "笑问客从何处来的上一句"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n笑问客从何处来的上一句 ⇔ ", "response": "来自天的笑问客何处", "predicted": -100, "true_label": "李白写了多少诗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这两个人即将在水球锦标赛中得分。 ⇔ ", "response": "这两个人即将在水球锦标赛中得分。", "predicted": -100, "true_label": "这两个人将要在一场比赛中竞争。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这两个人将要在一场比赛中竞争。 ⇔ ", "response": "红色汽车在绿色汽车的后面。", "predicted": -100, "true_label": "这两个人即将在水球锦标赛中得分。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nLETSGO ⇔ ", "response": "Letgo", "predicted": -100, "true_label": "Letmego"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nLetmego ⇔ ", "response": "LETMEGO", "predicted": -100, "true_label": "LETSGO"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is waking up ⇔ ", "response": "A girl is waking up comfortably", "predicted": -100, "true_label": "A child is waking up"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child is waking up ⇔ ", "response": "A child is falling back to sleep in bed", "predicted": -100, "true_label": "A girl is waking up"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n其他欧洲大国不再准备接受新的法国扩张，并愿意组建联盟抵制这种事情。 ⇔ ", "response": "法国试图在比利时建立殖民地，但其他欧洲大国不再准备接受新的法国扩张，并愿意联盟以抵制这种行为。", "predicted": -100, "true_label": "其他欧洲大国不再接受新的法国扩张，并准备组成联盟反对这样的事情。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n其他欧洲大国不再接受新的法国扩张，并准备组成联盟反对这样的事情。 ⇔ ", "response": "一个欧洲大国，对法国的扩张感到厌倦了，准备打破僵局，并准备另一个联盟，以反对这样的扩张。", "predicted": -100, "true_label": "其他欧洲大国不再准备接受新的法国扩张，并愿意组建联盟抵制这种事情。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我随时都在等待着您的，也就是说，等待令堂和令妹到来拉斯科利尼科夫稍动了动，想说什么；他的脸上露出激动不安的神情。 ⇔ ", "response": "我随时都在等待着您的，也就是说，等待令堂和令妹到来拉斯科利尼科夫稍动了动，想说什么；他的脸上露出激动不安的神情。", "predicted": -100, "true_label": "拉斯柯尔尼科夫动了动，好像要开口说话，他脸上现出略为激动的样子。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n拉斯柯尔尼科夫动了动，好像要开口说话，他脸上现出略为激动的样子。 ⇔ ", "response": "拉斯柯尔尼科夫动了动，好像要开口说话，他脸上现出略为激动的样子。", "predicted": -100, "true_label": "我随时都在等待着您的，也就是说，等待令堂和令妹到来拉斯科利尼科夫稍动了动，想说什么；他的脸上露出激动不安的神情。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is cutting cooked octopus ⇔ ", "response": "A woman is cutting a piece of octopus", "predicted": -100, "true_label": "A woman is cutting an octopus"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is cutting an octopus ⇔ ", "response": "A woman is cutting meat", "predicted": -100, "true_label": "The woman is cutting cooked octopus"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCLASSROOM ⇔ ", "response": "Room", "predicted": -100, "true_label": "Whatsinyourclassroom"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhatsinyourclassroom ⇔ ", "response": "Classroomitems", "predicted": -100, "true_label": "CLASSROOM"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young boy in a cowboy hat rides a large horse. ⇔ ", "response": "A boy on a horse.", "predicted": -100, "true_label": "A boy is pretending to be a cowboy."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy is pretending to be a cowboy. ⇔ ", "response": "A boy wearing jeans and a hat is riding in a pretend train.", "predicted": -100, "true_label": "A young boy in a cowboy hat rides a large horse."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这个是什么水果 ⇔ ", "response": "请问这是水果吗？", "predicted": -100, "true_label": "请问，这是什么水果？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问，这是什么水果？ ⇔ ", "response": "这是什么语言？", "predicted": -100, "true_label": "请问这个是什么水果"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nlsquo ⇔ ", "response": "lsqb", "predicted": -100, "true_label": "oppor9splusota"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor9splusota ⇔ ", "response": "OPPO r9s plus", "predicted": 9, "true_label": "lsquo"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman in black reviews a message as she walks to work. ⇔ ", "response": "The woman is trying to read the message on her phone.", "predicted": -100, "true_label": "The woman in black has a job."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman in black has a job. ⇔ ", "response": "One woman in a black suit and tie-dye shirt is looking at a computer.", "predicted": -100, "true_label": "A woman in black reviews a message as she walks to work."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真不索 ⇔ ", "response": "今天天气真好", "predicted": -100, "true_label": "今天天气真好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好 ⇔ ", "response": "今天天气真123", "predicted": 123, "true_label": "今天天气真不索"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man with a hard hat dancing ⇔ ", "response": "A man with a hard hat is dancing", "predicted": -100, "true_label": "A man with a hard hat is dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man with a hard hat is dancing ⇔ ", "response": "There is no man with a hard hat dancing", "predicted": -100, "true_label": "There is no man with a hard hat dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他取代约翰马布里作为德里克李的第一垒的替补。 ⇔ ", "response": "他替换了约翰·马布里，作为德里克李的第一垒的替补。", "predicted": -100, "true_label": "他取代约翰马布里作为第一基地的替补德里克李。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他取代约翰马布里作为第一基地的替补德里克李。 ⇔ ", "response": "他被交易到菲尼克斯，取代约翰马布里作为第一基地的替补德里克李。", "predicted": -100, "true_label": "他取代约翰马布里作为德里克李的第一垒的替补。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国最高政治顾问会见希腊总理 ⇔ ", "response": "希腊总理举行正式会谈", "predicted": -100, "true_label": "中国高级政治顾问会见也门总统"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国高级政治顾问会见也门总统 ⇔ ", "response": "杨利伟返回地球", "predicted": -100, "true_label": "中国最高政治顾问会见希腊总理"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAntiemetic drugs are drugs used to combat nausea and vomiting. ⇔ ", "response": "Drugs to combat nausea and vomiting are antiemetic drugs.", "predicted": -100, "true_label": "Drugs are used to combat schizophrenia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDrugs are used to combat schizophrenia. ⇔ ", "response": "Antipsychotics drugs are mainly used for the control of severe psychotic states such as schizophrenia and mania.", "predicted": -100, "true_label": "Antiemetic drugs are drugs used to combat nausea and vomiting."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气很漂亮 ⇔ ", "response": "我今天天很漂亮", "predicted": -100, "true_label": "我今天天气很漂流"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气很漂流 ⇔ ", "response": "我今天天气很好", "predicted": -100, "true_label": "我今天天气很漂亮"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n人民币兑美元汇率周一跌至6.3282。 ⇔ ", "response": "人民币兑美元汇率周一从周一早盘的高点下跌0.1%，至6.3195。", "predicted": 0, "true_label": "周四人民币兑美元汇率回落至6.2697。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n周四人民币兑美元汇率回落至6.2697。 ⇔ ", "response": "周四人民币兑美元汇率中间价升至6.2350。", "predicted": 6, "true_label": "人民币兑美元汇率周一跌至6.3282。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man riding a mechanical bull ⇔ ", "response": "A man is riding a mechanical bull", "predicted": -100, "true_label": "The man is riding a mechanical bull"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is riding a mechanical bull ⇔ ", "response": "The man is riding a pony", "predicted": -100, "true_label": "There is no man riding a mechanical bull"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n十亿。世界上的每一个人都达到了墨西哥目前的繁荣水平。 ⇔ ", "response": "如果世界上的每一人达到墨西哥目前的繁荣水平，会是多少？", "predicted": -100, "true_label": "世界上的每一个人都处在目前西北非洲的“繁荣”水平。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n世界上的每一个人都处在目前西北非洲的“繁荣”水平。 ⇔ ", "response": ")尼及利亚人可以期望政府给他们提供良好的服务，从卫生到交通部门。", "predicted": -100, "true_label": "十亿。世界上的每一个人都达到了墨西哥目前的繁荣水平。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl in a black sweater and jeans pours water into an empty soda bottle. ⇔ ", "response": "The girl is filling a soda bottle.", "predicted": -100, "true_label": "Someone pours water into a bottle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone pours water into a bottle ⇔ ", "response": "Water is poured into a plastic bottle by a worker", "predicted": -100, "true_label": "A girl in a black sweater and jeans pours water into an empty soda bottle."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nState Supreme Court Justice Ira Gammerman said in court this morning , \" It was an ill-conceived lawsuit . \" ⇔ ", "response": "State Supreme Court Justice Yisroel Gammerman said in open court today , he considered the lawsuit ill-conceived .", "predicted": -100, "true_label": "New York Supreme Court Justice Ira Gammerman said in his statement that the lawsuit was \" ill-conceived . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNew York Supreme Court Justice Ira Gammerman said in his statement that the lawsuit was \" ill-conceived . \" ⇔ ", "response": "New York Supreme Court Justice Ira Gammerman said the lawsuit was an \" ill-conceived\" use of legal resources .", "predicted": -100, "true_label": "State Supreme Court Justice Ira Gammerman said in court this morning , \" It was an ill-conceived lawsuit . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man draws a caricature of a woman. ⇔ ", "response": "A man is drawing something; the background is unclear.", "predicted": -100, "true_label": "A man drawing a caricature of his wife."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man drawing a caricature of his wife. ⇔ ", "response": "A man is drawing a picture of his wife.", "predicted": -100, "true_label": "A man draws a caricature of a woman."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这种植物叫什么名字 ⇔ ", "response": "请问这种花叫什么名字", "predicted": -100, "true_label": "请问这种植物叫什么名字、"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这种植物叫什么名字、 ⇔ ", "response": "请问这个叫什么？", "predicted": -100, "true_label": "请问这种植物叫什么名字"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n官方SpA新闻社发表的一份声明称，al-Dosari被判犯有贩毒罪。 ⇔ ", "response": "官方SpA新闻社(AA)表示，萨那法院判定al-Dosari犯有贩毒罪。", "predicted": -100, "true_label": "强奸犯被确认为HosseinSafarlou，并被判定犯有贩毒罪。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n强奸犯被确认为HosseinSafarlou，并被判定犯有贩毒罪。 ⇔ ", "response": "HosseinSafarlou被判定犯有强奸罪并定为贩毒犯。", "predicted": -100, "true_label": "官方SpA新闻社发表的一份声明称，al-Dosari被判犯有贩毒罪。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAki Kaurismaki was born in Orimattila, Finland, on 4th April, 1957. After studying film, he wrote, produced and directed Saimaailmio (1981) in Finland. ⇔ ", "response": "Aki Kaurismaki was born in Orimattila, Finland.", "predicted": -100, "true_label": "Aki Kaurismaki directed a film."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAki Kaurismaki directed a film. ⇔ ", "response": "Finnish director Aki Kaurismaki is shooting his new film, about missing girls, in dark rooms and low light.", "predicted": -100, "true_label": "Aki Kaurismaki was born in Orimattila, Finland, on 4th April, 1957. After studying film, he wrote, produced and directed Saimaailmio (1981) in Finland."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n关于友谊的名人名言 ⇔ ", "response": "名人名言栏目下有一篇关于友谊的名人名言", "predicted": -100, "true_label": "有关天气的谚语"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有关天气的谚语 ⇔ ", "response": "天气对体育比赛的影响", "predicted": -100, "true_label": "关于友谊的名人名言"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你们现在都在干嘛。 ⇔ ", "response": "你在干什么", "predicted": -100, "true_label": "现在都在干嘛呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n现在都在干嘛呢 ⇔ ", "response": "你在干什么呢哦", "predicted": -100, "true_label": "你们现在都在干嘛。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no young boy playing a wind instrument ⇔ ", "response": "A young boy is playing a wind instrument", "predicted": -100, "true_label": "A young boy is playing a wind instrument"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young boy is playing a wind instrument ⇔ ", "response": "A boy is playing a wind instrument", "predicted": -100, "true_label": "There is no young boy playing a wind instrument"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDid the media pounce upon the religion of Jurgen Schneider, the developer who went bankrupt and fled Germany owing billions of DMs to Deutsche Bank and others, as the cause of his downfall? ⇔ ", "response": "Jurgen Schneider went bankrupt and fled Germany.", "predicted": -100, "true_label": "Jurgen Schneider went bankrupt."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nJurgen Schneider went bankrupt. ⇔ ", "response": "The German coach, who last week accused his players of not caring enough to win, went bankrupt yesterday.", "predicted": -100, "true_label": "Did the media pounce upon the religion of Jurgen Schneider, the developer who went bankrupt and fled Germany owing billions of DMs to Deutsche Bank and others, as the cause of his downfall?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你挺好的啊 ⇔ ", "response": "你觉得挺好的", "predicted": -100, "true_label": "你挺能干的话"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你挺能干的话 ⇔ ", "response": "你能干什么啊", "predicted": -100, "true_label": "你挺好的啊"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAhold Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for the scandal . ⇔ ", "response": "In front of more than 500 angry shareholders , Ahold chief executive Henny de Ruiter offered his \" sincere apologies \" for the scandal , which has forced him to step down .", "predicted": 500, "true_label": "Board Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for events at Foodservice and elsewhere in the embattled group ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBoard Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for events at Foodservice and elsewhere in the embattled group . ⇔ ", "response": "Henny de Ruiter , board chairman , offered his \" sincere apologies \" for events at Foodservice and elsewhere in the embattled group .", "predicted": -100, "true_label": "Ahold Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for the scandal ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA picture is being drawn by a man ⇔ ", "response": "A man is drawing a picture", "predicted": -100, "true_label": "The person is drawing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe person is drawing ⇔ ", "response": "A man is drawing", "predicted": -100, "true_label": "A picture is being drawn by a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree dogs running through a field. ⇔ ", "response": "Three dogs playing in a field.", "predicted": -100, "true_label": "Three dogs are running"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree dogs are running ⇔ ", "response": "Three dogs are running in the park", "predicted": -100, "true_label": "Three dogs running through a field."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man with a paddle stands next to a woman with a bikini top; and her arms are in a dance pose. ⇔ ", "response": "A man with a paddle is at the shore.", "predicted": -100, "true_label": "someone dances while a man holds a paddle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nsomeone dances while a man holds a paddle ⇔ ", "response": "A man holds a paddle and dances.", "predicted": -100, "true_label": "A man with a paddle stands next to a woman with a bikini top; and her arms are in a dance pose."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我想玩抖音 ⇔ ", "response": "抖音里面有句话怎么样", "predicted": -100, "true_label": "继续播放抖音"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n继续播放抖音 ⇔ ", "response": "抖音怎么关闭", "predicted": -100, "true_label": "我想玩抖音"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is wearing a hard hat and dancing ⇔ ", "response": "There is no man wearing a hard hat and dancing", "predicted": -100, "true_label": "There is no man with a hard hat dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man with a hard hat dancing ⇔ ", "response": "A man with a hard hat is dancing", "predicted": -100, "true_label": "A man is wearing a hard hat and dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe CFAP purchases food stamp coupons from the federal government and distributes them to eligible recipients. ⇔ ", "response": "People receive food stamps from the CFAP.", "predicted": -100, "true_label": "A government purchases food."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA government purchases food. ⇔ ", "response": "Restricted items are those items that are not acceptable for purchase using government appropriated funds. Some examples of items that the General Accounting Office (GAO) considers \"restricted\" include: food, live entertainment, and personal gifts including plaques.", "predicted": -100, "true_label": "The CFAP purchases food stamp coupons from the federal government and distributes them to eligible recipients."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n北京哪里租房子最便宜？ ⇔ ", "response": "北京租房价格最便宜的区是哪里？", "predicted": -100, "true_label": "北京哪里租房子最便宜？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n北京哪里租房子最便宜？ ⇔ ", "response": "北京租房价格最便宜的区是哪里？", "predicted": -100, "true_label": "北京哪里租房子最便宜？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我想给我的饭店起个名字 ⇔ ", "response": "我想要个饭店的名字", "predicted": -100, "true_label": "我想给小饭店起个名字，"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我想给小饭店起个名字， ⇔ ", "response": "我想给小饭店起个名字吧123456789", "predicted": 123456789, "true_label": "我想给我的饭店起个名字"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在狭窄的小路上骑自行车的人。 ⇔ ", "response": "一名男子骑着自行车在狭窄的小路上行驶。", "predicted": -100, "true_label": "一名男子骑着独轮车沿着一条铺好的道路行驶。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一名男子骑着独轮车沿着一条铺好的道路行驶。 ⇔ ", "response": "一名男子正用独轮车沿着一条铺好的道路行驶。", "predicted": -100, "true_label": "在狭窄的小路上骑自行车的人。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is playing the guitar ⇔ ", "response": "There is no man playing the guitar", "predicted": -100, "true_label": "Nobody is playing the guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNobody is playing the guitar ⇔ ", "response": "A man is playing the guitar", "predicted": -100, "true_label": "A man is playing the guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天昨了个美梦 ⇔ ", "response": "我昨天做一个美梦", "predicted": -100, "true_label": "我昨天做了个美梦"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天做了个美梦 ⇔ ", "response": "我昨天做了个噩梦", "predicted": -100, "true_label": "我昨天昨了个美梦"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天会下雨吗？ ⇔ ", "response": "明天会下雨吗？", "predicted": -100, "true_label": "明天会下雨马？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天会下雨马？ ⇔ ", "response": "明天会下雨吗？", "predicted": -100, "true_label": "明天会下雨吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTony winners will be announced June 8 at the Radio City Music Hall in New York . ⇔ ", "response": "The Tony Awards honoring the 2010 main award nominees will be held on June 8 at Radio City Music Hall in New York City .", "predicted": 2010, "true_label": "Winners will be announced in a June 8 ceremony broadcast on CBS from Radio City Music Hall ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWinners will be announced in a June 8 ceremony broadcast on CBS from Radio City Music Hall . ⇔ ", "response": "Winners will be announced at a June 8 ceremony at Radio City Music Hall broadcast on CBS .", "predicted": 8, "true_label": "Tony winners will be announced June 8 at the Radio City Music Hall in New York ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe animal with big eyes is voraciously eating ⇔ ", "response": "The animal with big eyes is eating a leaf", "predicted": -100, "true_label": "The animal with big eyes is eating"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe animal with big eyes is eating ⇔ ", "response": "A lemur is eating quickly", "predicted": -100, "true_label": "The animal with big eyes is voraciously eating"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA female runner dressed in blue athletic wear is running in a competition, while spectators line the street. ⇔ ", "response": "A runner is competing against the other runners.", "predicted": -100, "true_label": "There are people on the street."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are people on the street. ⇔ ", "response": "A person in a black shirt walks down the street past other people who are wearing clothes from different stores.", "predicted": -100, "true_label": "A female runner dressed in blue athletic wear is running in a competition, while spectators line the street."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nwhatyou ⇔ ", "response": "WHATYOU", "predicted": -100, "true_label": "withoutyou"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nwithoutyou ⇔ ", "response": "lemounfa", "predicted": -100, "true_label": "whatyou"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman playing an electric guitar ⇔ ", "response": "A woman is playing an electric guitar", "predicted": -100, "true_label": "A woman is playing an electric guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is playing an electric guitar ⇔ ", "response": "A woman is playing an acoustic guitar", "predicted": -100, "true_label": "There is no woman playing an electric guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n呼叫爸爸呼叫爸爸 ⇔ ", "response": "打电话给爸爸", "predicted": -100, "true_label": "呼叫起来"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n呼叫起来 ⇔ ", "response": "打电话给", "predicted": -100, "true_label": "呼叫爸爸呼叫爸爸"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n求这个动态图的出处 ⇔ ", "response": "动态图出处", "predicted": -100, "true_label": "动态图，求出处。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n动态图，求出处。 ⇔ ", "response": "动态图是什么。", "predicted": -100, "true_label": "求这个动态图的出处"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这完全取决于你的公司和你的合同。 ⇔ ", "response": "这取决于你的国家和合同。", "predicted": -100, "true_label": "那要看你的公司了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n那要看你的公司了。 ⇔ ", "response": "那要视你的公司而定。", "predicted": -100, "true_label": "这完全取决于你的公司和你的合同。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第三方理财公司可信吗? ⇔ ", "response": "第三方理财公司靠不靠谱？", "predicted": -100, "true_label": "投资理财公司可信吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n投资理财公司可信吗？ ⇔ ", "response": "信托公司可信吗", "predicted": -100, "true_label": "第三方理财公司可信吗?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoR15 ⇔ ", "response": "opppoR15Xtog", "predicted": 15, "true_label": "oppr15"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppr15 ⇔ ", "response": "r11p", "predicted": 11, "true_label": "oppoR15"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe petition alleges that Huletts unfair sales have damaged the U.S. industry , sending market prices below sustainable levels . ⇔ ", "response": "The petition alleges that Huletts has engaged in unfair sales practices in the U.S. shrimp industry, sending market prices below sustainable levels.", "predicted": -100, "true_label": "Those unfair sales have damaged the US industry by eroding market prices below sustainable levels , says Alcoa ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThose unfair sales have damaged the US industry by eroding market prices below sustainable levels , says Alcoa . ⇔ ", "response": "The company said such sales had damaged the U.S. industry by eroding market prices below sustainable levels .", "predicted": -100, "true_label": "The petition alleges that Huletts unfair sales have damaged the U.S. industry , sending market prices below sustainable levels ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nREADING ⇔ ", "response": "British", "predicted": -100, "true_label": "REVIEW"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nREVIEW ⇔ ", "response": "See you", "predicted": -100, "true_label": "READING"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young child joyfully pulls colorful tissue paper from a decorated box, looking for his present. ⇔ ", "response": "a child opens a present on his birthday", "predicted": -100, "true_label": "a child pulls colorful tissue paper from a fancy box"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na child pulls colorful tissue paper from a fancy box ⇔ ", "response": "A child pulls colorful paper from a box.", "predicted": -100, "true_label": "A young child joyfully pulls colorful tissue paper from a decorated box, looking for his present."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n五音有什么 ⇔ ", "response": "五音各是什么", "predicted": -100, "true_label": "零点二五等于多少"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n零点二五等于多少 ⇔ ", "response": "零点二五和二点五一样吗", "predicted": -100, "true_label": "五音有什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样调理痛经呢？ ⇔ ", "response": "痛经怎么调理？", "predicted": -100, "true_label": "痛经怎么调理，吃什么好？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n痛经怎么调理，吃什么好？ ⇔ ", "response": "痛经怎么调理，吃什么好？", "predicted": -100, "true_label": "怎样调理痛经呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n帮我打开美颜自拍 ⇔ ", "response": "能不能用我的手机拍照片", "predicted": -100, "true_label": "给我拍照"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n给我拍照 ⇔ ", "response": "我想要拍照", "predicted": -100, "true_label": "帮我打开美颜自拍"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBoy in midair on bicycle ⇔ ", "response": "A boy in an orange uniform lands from a great height on a metal bike rack.", "predicted": -100, "true_label": "A boy is airborne on a bicycle."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy is airborne on a bicycle. ⇔ ", "response": "A boy on a bicycle in the air.", "predicted": -100, "true_label": "Boy in midair on bicycle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n伊朗近几个月来明显增加了处决人数，因为伊朗官员正在实施一项旨在促进道德和加强伊朗安全的运动。 ⇔ ", "response": "伊朗lower February处决人数表明，伊朗官员正在逐步实施一项运动，促进道德和加强伊朗安全。", "predicted": -100, "true_label": "伊朗官员在2007年增加了处决人数，因为伊朗官员实施了一场据称旨在促进道德和加强社会安全的运动。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n伊朗官员在2007年增加了处决人数，因为伊朗官员实施了一场据称旨在促进道德和加强社会安全的运动。 ⇔ ", "response": "2007年，伊朗官员因所谓的有组织犯罪处决更多的人，官员们声称这旨在加强社会安全并促进道德。", "predicted": 2007, "true_label": "伊朗近几个月来明显增加了处决人数，因为伊朗官员正在实施一项旨在促进道德和加强伊朗安全的运动。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么软件看电影最好。 ⇔ ", "response": "什么网页看电影最好。", "predicted": -100, "true_label": "看电影用什么软件最好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n看电影用什么软件最好 ⇔ ", "response": "软件安装用什么软件最好", "predicted": -100, "true_label": "什么软件看电影最好。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。 ⇔ ", "response": "音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。", "predicted": -100, "true_label": "音乐由A. T. Ummer编写，歌词由Koorkkancheri Sugathan和Poovachal Khader创作。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n音乐由A. T. Ummer编写，歌词由Koorkkancheri Sugathan和Poovachal Khader创作。 ⇔ ", "response": "音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。", "predicted": -100, "true_label": "音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is slowly dancing on the road ⇔ ", "response": "A man is dancing on the road", "predicted": -100, "true_label": "A man is dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is dancing ⇔ ", "response": "There is no man dancing", "predicted": -100, "true_label": "A man is slowly dancing on the road"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMan jumping up in the air and splitting his legs with his mouth open. ⇔ ", "response": "A man is jumping.", "predicted": -100, "true_label": "A man is jumping."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is jumping. ⇔ ", "response": "A man does a flip over a boulder in the sun.", "predicted": -100, "true_label": "Man jumping up in the air and splitting his legs with his mouth open."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIran has yet to sign an additional protocol to the NPT treaty which would allow U.N. inspections at short notice . ⇔ ", "response": "But Iran has not signed a supplementary protocol that would allow U.N. inspections at short notice .", "predicted": -100, "true_label": "Iran has yet to sign an additional protocol to the Nuclear Non-Proliferation Treaty , which it signed in 1970 , that would allow IAEA inspections at short notice ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIran has yet to sign an additional protocol to the Nuclear Non-Proliferation Treaty , which it signed in 1970 , that would allow IAEA inspections at short notice . ⇔ ", "response": "Iran has not signed a supplementary protocol to the Nuclear Non-Proliferation Treaty, which it signed in 1970, that would allow for inspections at short notice .", "predicted": 1970, "true_label": "Iran has yet to sign an additional protocol to the NPT treaty which would allow U.N. inspections at short notice ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样开通农行手机银行 ⇔ ", "response": "如何办理农行手机银行", "predicted": -100, "true_label": "怎样开通农行手机银行？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样开通农行手机银行？ ⇔ ", "response": "阿农怎么开通手机银行了", "predicted": -100, "true_label": "怎样开通农行手机银行"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSoccer players are kicking a soccer ball into the goal ⇔ ", "response": "A player of soccer is scoring a goal", "predicted": -100, "true_label": "There are no soccer players kicking a soccer ball into the goal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are no soccer players kicking a soccer ball into the goal ⇔ ", "response": "A soccer player is kicking a soccer ball into the goal", "predicted": -100, "true_label": "Soccer players are kicking a soccer ball into the goal"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国联通中国移动中国电信哪个好? ⇔ ", "response": "中国电信和中国联通哪个好？", "predicted": -100, "true_label": "中国电信中国移动中国联通上网哪个好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国电信中国移动中国联通上网哪个好 ⇔ ", "response": "移动和中国电信哪个网好", "predicted": -100, "true_label": "中国联通中国移动中国电信哪个好?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n来自世界各地的编辑敦促土耳其保护媒体 ⇔ ", "response": "土耳其必须保护自由媒体，来自世界各地的编辑写信说", "predicted": -100, "true_label": "土耳其打击反政府武装，媒体自由岌岌可危"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n土耳其打击反政府武装，媒体自由岌岌可危 ⇔ ", "response": "土耳其加强媒体自由，媒体自由岌岌可危", "predicted": -100, "true_label": "来自世界各地的编辑敦促土耳其保护媒体"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThis model allows for different levels of training to be offered to the various skill levels of soldiers that will be operating and maintaining a particular system. ⇔ ", "response": "Soldiers operate a system.", "predicted": -100, "true_label": "Soldiers are trained using virtual reality."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSoldiers are trained using virtual reality. ⇔ ", "response": "Virtual reality has many applications. Today the military uses this technology in simulations. Soldiers are trained to fly airplanes, drive tanks and fight wars without having to physically be in danger.", "predicted": -100, "true_label": "This model allows for different levels of training to be offered to the various skill levels of soldiers that will be operating and maintaining a particular system."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone is drawing a person ⇔ ", "response": "The person is being drawn by the pencil", "predicted": -100, "true_label": "The person is drawing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe person is drawing ⇔ ", "response": "A man is drawing", "predicted": -100, "true_label": "Someone is drawing a person"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n吃完螃蟹能吃橘子吗 ⇔ ", "response": "吃完螃蟹不能吃苹果吗", "predicted": -100, "true_label": "鱼吃什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n鱼吃什么 ⇔ ", "response": "鱼吃什么食物", "predicted": -100, "true_label": "吃完螃蟹能吃橘子吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n课一结束，骚动便随之而来，但她的话音刚落，全校又复归平静，她继续说:今天早晨的早饭，你们都吃不下去，大家一定饿坏了，我己经吩咐给大家准备了面包和乳酪当点心，教师们带着某种惊异的目光看着她。 ⇔ ", "response": "她的话音刚落，全校又复归平静。", "predicted": -100, "true_label": "她接着往下说:今天早上你们早饭吃不下去；现在一定都饿了；——我已经吩咐过，给大伙儿准备一顿面包和干酪的点心。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她接着往下说:今天早上你们早饭吃不下去；现在一定都饿了；——我已经吩咐过，给大伙儿准备一顿面包和干酪的点心。 ⇔ ", "response": "她接着往下说：“今天早上你们早饭吃不下去；现在一定都饿了；——我已经吩咐过，给大伙儿准备一顿面包和干酪的点心。", "predicted": -100, "true_label": "课一结束，骚动便随之而来，但她的话音刚落，全校又复归平静，她继续说:今天早晨的早饭，你们都吃不下去，大家一���饿坏了，我己经吩咐给大家准备了面包和乳酪当点心，教师们带着某种惊异的目光看着她。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn woman that is old is shaking hands with a man ⇔ ", "response": "An old woman is shaking hands with a young man", "predicted": -100, "true_label": "A man and a woman are shaking hands"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are shaking hands ⇔ ", "response": "A man and a woman are standing near each other and shaking hands", "predicted": -100, "true_label": "An woman that is old is shaking hands with a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMan holding torch of fire. ⇔ ", "response": "A man is holding a torch of fire.", "predicted": -100, "true_label": "There is a man."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is a man. ⇔ ", "response": "A man in a black suit and a woman in a white dress walk along a sidewalk.", "predicted": -100, "true_label": "Man holding torch of fire."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is applying makeup to her face ⇔ ", "response": "There is no girl applying makeup to her face", "predicted": -100, "true_label": "There is no woman putting on makeup"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman putting on makeup ⇔ ", "response": "A woman is putting on makeup", "predicted": -100, "true_label": "A girl is applying makeup to her face"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn the early 20th century, agoraphobia, or the fear of public spaces, saw a transformation in its demographic composition. The disease's early history can be characterized as a predominantly male disorder where the majority of patients diagnosed with the disease were white, middle-class men. ⇔ ", "response": "Agoraphobia was predominantly a male disorder.", "predicted": -100, "true_label": "Agoraphobia is a widespread disorder."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAgoraphobia is a widespread disorder. ⇔ ", "response": "Although panic disorders and agoraphobia are widespread in the general population, almost half of those afflicted with these problems do not receive treatment for them.", "predicted": -100, "true_label": "In the early 20th century, agoraphobia, or the fear of public spaces, saw a transformation in its demographic composition. The disease's early history can be characterized as a predominantly male disorder where the majority of patients diagnosed with the disease were white, middle-class men."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n赛普拉斯县由MHCW的联邦选举部门服务，并由保守党议员GLEN MOTZ代表加拿大下议院。 ⇔ ", "response": "赛普拉斯县由MHCW的联邦选举部门服务，并由保守党议员GLEN MOTZ代表加拿大众议院。", "predicted": -100, "true_label": "赛普拉斯县由MHCW的联邦选举部门代表，并由保守党成员GLEN MOTZ在加拿大下议院任职。"}

eval_gen/gpu_results/results_gpu_3.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n赛普拉斯县由MHCW的联邦选举部门代表，并由保守党成员GLEN MOTZ在加拿大下议院任职。 ⇔ ", "response": "赛普拉斯县由MHCW的联邦选举部门代表，并由加拿大保守党成员GLEN MOTZ在加拿大下议院任职。", "predicted": -100, "true_label": "赛普拉斯县由MHCW的联邦选举部门服务，并由保守党议员GLEN MOTZ代表加拿大下议院。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机怎么扩展运行内存 ⇔ ", "response": "如何使手机的运行内存增大", "predicted": -100, "true_label": "怎么增加手机运行内存"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么增加手机运行内存 ⇔ ", "response": "如何使手机运行内存变大", "predicted": -100, "true_label": "手机怎么扩展运行内存"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo dirt bike riders, one wearing green and the other wearing blue and white, are jumping a hill. ⇔ ", "response": "Two people are riding dirt bikes.", "predicted": -100, "true_label": "Two dirt bike riders are outside"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo dirt bike riders are outside ⇔ ", "response": "Two riders are engaged in dirt biking", "predicted": -100, "true_label": "Two dirt bike riders, one wearing green and the other wearing blue and white, are jumping a hill."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe recent collision between a Super Ferry and a smaller ship that killed at least 38 people has once again put the spotlight on the Philippine Coast Guard. ⇔ ", "response": "The Philippine Coast Guard was involved in a collision.", "predicted": -100, "true_label": "People were killed in a collision between ships."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople were killed in a collision between ships. ⇔ ", "response": "On December 6, as a result of a collision between two ships in Halifax harbour, a cargo of ammunition exploded and 1,630 people lost their lives in the ensuing holocaust, thousands more being injured.", "predicted": 6, "true_label": "The recent collision between a Super Ferry and a smaller ship that killed at least 38 people has once again put the spotlight on the Philippine Coast Guard."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is phoning ⇔ ", "response": "There is no man on the phone", "predicted": -100, "true_label": "A man is talking on the phone"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is talking on the phone ⇔ ", "response": "A man is not talking on the phone", "predicted": -100, "true_label": "A man is phoning"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n很酷的动漫女孩图片 ⇔ ", "response": "动漫女孩很酷的图片", "predicted": -100, "true_label": "动漫可爱女孩的图片"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n动漫可爱女孩的图片 ⇔ ", "response": "请求动漫女孩图片", "predicted": -100, "true_label": "很酷的动漫女孩图片"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人在割花。 ⇔ ", "response": "一个女人在割芦苇。", "predicted": -100, "true_label": "一个女人在切西兰花。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人在切西兰花。 ⇔ ", "response": "一个女人在切西葫芦。", "predicted": -100, "true_label": "一个女人在割花。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n甲沟炎怎么在家自己治疗 ⇔ ", "response": "甲沟炎怎么治疗", "predicted": -100, "true_label": "甲沟炎怎么治疗【甲沟炎】"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n甲沟炎怎么治疗【甲沟炎】 ⇔ ", "response": "甲沟炎怎么办", "predicted": -100, "true_label": "甲沟炎怎么在家自己治疗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了苹果和 ⇔ ", "response": "我今天吃了苹果和香蕉", "predicted": -100, "true_label": "今天早上吃了苹果和牛奶"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天早上吃了苹果和牛奶 ⇔ ", "response": "今天早上吃了苹果和巴菲", "predicted": -100, "true_label": "我今天吃了苹果和"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气晴朗，我去公园散步。 ⇔ ", "response": "今天天气晴朗，我去公园散步了。", "predicted": -100, "true_label": "今天天气晴朗，我去公园伞步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气晴朗，我去公园伞步。 ⇔ ", "response": "今天天气晴朗，我去公园散步。", "predicted": -100, "true_label": "今天天气晴朗，我去公园散步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n觉得生活没意思咋办 ⇔ ", "response": "觉得生活没意义怎么办", "predicted": -100, "true_label": "有时候感觉生活真没意思"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有时候感觉生活真没意思 ⇔ ", "response": "有时候感觉生活有意思", "predicted": -100, "true_label": "觉得生活没意思咋办"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n被斯蒂芬·曼根（Stephen Mangan）阅读的佩尼·莱斯特（Penny Leicester）删除，彼得·塞拉芬诺维奇（Peter Serafinowicz）作为“指南”（The Guide）的声音，由希瑟·拉莫尔（Heather Larmour）制作。 ⇔ ", "response": "被斯蒂芬·曼根（Stephen Mangan）阅读的佩尼·莱斯特（Penny Leicester）删除，彼得·塞拉芬诺维奇（Peter Serafinowicz）作为“指南”（The Guide）的声音，由希瑟·拉莫尔（Heather Larmour）制作。", "predicted": -100, "true_label": "它由Penny Leicester缩短，由Stephen Mangan阅读，由Peter Serafinowicz作为导游的声音和Heather Larmour制作。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它由Penny Leicester缩短，由Stephen Mangan阅读，由Peter Serafinowicz作为导游的声音和Heather Larmour制作。 ⇔ ", "response": "它由Stephen Mangan缩短，由Peter Serafinowicz作为导游的声音阅读，由Heather Larmour制作。", "predicted": -100, "true_label": "被斯蒂芬·曼根（Stephen Mangan）阅读的佩尼·莱斯特（Penny Leicester）删除，彼得·塞拉芬诺维奇（Peter Serafinowicz）作为“指南”（The Guide）的声音，由希瑟·拉莫尔（Heather Larmour）制作。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPO的设计理念是 ⇔ ", "response": "OPPO的设计理念是什么？", "predicted": -100, "true_label": "善良都是个性的正面"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n善良都是个性的正面 ⇔ ", "response": "性格要独特", "predicted": -100, "true_label": "OPPO的设计理念是"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe young boy in a red shirt and jeans resists his sister's request for water on the road. ⇔ ", "response": "A young boy is resisting his sister's request for water.", "predicted": -100, "true_label": "Kids don't share water."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nKids don't share water. ⇔ ", "response": "Two kids get water from the same bottle.", "predicted": -100, "true_label": "The young boy in a red shirt and jeans resists his sister's request for water on the road."}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoA ⇔ ", "response": "oppooppoA9X", "predicted": 9, "true_label": "oppoa79ota"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoa79ota ⇔ ", "response": "OPPOa4", "predicted": 4, "true_label": "oppoA"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people play horseshoes. ⇔ ", "response": "Two players in a game of horseshoes.", "predicted": -100, "true_label": "Tom and Sue were playing outside."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTom and Sue were playing outside. ⇔ ", "response": "Tom played soccer with his friends Sue and John.", "predicted": -100, "true_label": "Two people play horseshoes."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n求现代军事题材电视剧 ⇔ ", "response": "求现代军事题材的电视剧名", "predicted": -100, "true_label": "国产现代军事题材的电视剧"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n国产现代军事题材的电视剧 ⇔ ", "response": "电视剧现代军旅题材国产", "predicted": -100, "true_label": "求现代军事题材电视剧"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe United States has threatened to end U.N. peacekeeping in Bosnia unless American personnel are given immunity from prosecution by the new international war-crimes tribunal. ⇔ ", "response": "The United States pulls out of Bosnia.", "predicted": -100, "true_label": "U.N. peacekeeping forces withdrew from Bosnia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nU.N. peacekeeping forces withdrew from Bosnia. ⇔ ", "response": "The decision by the U.N. Security Council to end the peacekeeping force in Bosnia comes as a surprise even to those most familiar with the crisis.", "predicted": -100, "true_label": "The United States has threatened to end U.N. peacekeeping in Bosnia unless American personnel are given immunity from prosecution by the new international war-crimes tribunal."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo men are fighting ⇔ ", "response": "Two men are grappling with each other", "predicted": -100, "true_label": "Some men are fighting"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome men are fighting ⇔ ", "response": "Two men are fighting", "predicted": -100, "true_label": "Two men are fighting"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n主屏幕时间怎样没有显示天气 ⇔ ", "response": "我的主屏幕时间没有显示天气", "predicted": -100, "true_label": "怎样在屏幕上显示时间和天气预报"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样在屏幕上显示时间和天气预报 ⇔ ", "response": "怎样在屏幕上显示字母表", "predicted": -100, "true_label": "主屏幕时间怎样没有显示天气"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCloud9是BeagleBone黑色单板计算机的原生IDE，主要在node.js的扩展中编程，称为Bonescript。 ⇔ ", "response": "Cloud9是BeagleBone黑色单板计算机的原生IDE，主要在node.js的扩展中编程，称为Bonescript。", "predicted": 9, "true_label": "Cloud9是本机板载计算机BeagleBone Black的唯一IDE，它主要在node.js的扩展中编程，称为Bonescript。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCloud9是本机板载计算机BeagleBone Black的唯一IDE，它主要在node.js的扩展中编程，称为Bonescript。 ⇔ ", "response": "Cloud9是本机板载计算机BeagleBone Black的唯一IDE，它主要在node.js中的扩展Bonescript中编程，称为扩展。", "predicted": 9, "true_label": "Cloud9是BeagleBone黑色单板计算机的原生IDE，主要在node.js的扩展中编程，称为Bonescript。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n设置输入法皮肤设置输入法皮肤 ⇔ ", "response": "设置输入法图标", "predicted": -100, "true_label": "输入法设置"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n输入法设置 ⇔ ", "response": "设置输入法", "predicted": -100, "true_label": "设置输入法皮肤设置输入法皮肤"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n丑是我的代名词。 ⇔ ", "response": "我喜欢看电影。", "predicted": -100, "true_label": "丑是儿子的代名词。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n丑是儿子的代名词。 ⇔ ", "response": "儿子叫丑是。", "predicted": -100, "true_label": "丑是我的代名词。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n通过两个通道上支持的数据，可以复制冗余通信。 ⇔ ", "response": "复制冗余通信在接收通道上支持数据。", "predicted": -100, "true_label": "两个通道上的复制数据都支持冗余通信。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两个通道上的复制数据都支持冗余通信。 ⇔ ", "response": "两个通道上的复制数据支持冗余通信。", "predicted": -100, "true_label": "通过两个通道上支持的数据，可以复制冗余通信。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的手机卡开关开不了 ⇔ ", "response": "我的手机卡关开不了", "predicted": -100, "true_label": "我的手机卡有流量吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的手机卡有流量吗 ⇔ ", "response": "我的手机流量用了吗", "predicted": -100, "true_label": "我的手机卡开关开不了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在那里，Anna Wallner遇到了Matisic并创造了“购物袋”概念。 ⇔ ", "response": "在那里，Anna Wallner遇到了Matisic，并创造了“购物袋”概念。", "predicted": -100, "true_label": "在那里，Matisic遇到了Anna Wallner并创造了“购物袋”的概念。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在那里，Matisic遇到了Anna Wallner并创造了“购物袋”的概念。 ⇔ ", "response": "在那里，Matisic遇到了Anna Wallner并创造了“购物袋”的概念。", "predicted": -100, "true_label": "在那里，Anna Wallner遇到了Matisic并创造了“购物袋”概念。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nYet another big fight has been over a White House proposal to privatize air-traffic control at major airports . ⇔ ", "response": "Yet another big fight has broken out over a White House plan to privatize air-traffic control at major airports .", "predicted": -100, "true_label": "Yet another fight has been waged over a White House proposal to privatize air traffic control at major airports , such as those in Pittsburgh and Philadelphia ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nYet another fight has been waged over a White House proposal to privatize air traffic control at major airports , such as those in Pittsburgh and Philadelphia . ⇔ ", "response": "Yet another fight has been waged over the White House proposal to privatize air traffic control at major airports such as those in Pittsburgh and Philadelphia .", "predicted": -100, "true_label": "Yet another big fight has been over a White House proposal to privatize air-traffic control at major airports ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n人生得意须尽欢下一句是啥 ⇔ ", "response": "下一句是哪一个", "predicted": -100, "true_label": "人生是啥"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n人生是啥 ⇔ ", "response": "人生 Meaning", "predicted": -100, "true_label": "人生得意须尽欢下一句是啥"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n纽约证券交易所(NewYorkStockExchange)下跌股与上涨股之比接近2：1。 ⇔ ", "response": "早上好，我的孩子，<|userkeyword|>\"他说，\"你今天怎么样？\"", "predicted": -100, "true_label": "纽约证交所上涨股与下跌股之比接近8：3，纳斯达克(Nasdaq)上涨股与下跌股之比超过11：5。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n纽约证交所上涨股与下跌股之比接近8：3，纳斯达克(Nasdaq)上涨股与下跌股之比超过11：5。 ⇔ ", "response": "纽约证交所与纳斯达克，升跌股比例", "predicted": -100, "true_label": "纽约证券交易所(NewYorkStockExchange)下跌股与上涨股之比接近2：1。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man on stage is singing into the microphone ⇔ ", "response": "A man in a suit is standing at a microphone and singing", "predicted": -100, "true_label": "The man on stage isn 't singing into the microphone"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man on stage isn 't singing into the microphone ⇔ ", "response": "The man on stage is singing into the microphone", "predicted": -100, "true_label": "The man on stage is singing into the microphone"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDefense lawyers had objected in pretrial hearings to the videotape and the photo , saying they were too unclear to identify Muhammad . ⇔ ", "response": "Defense lawyers had objected to the videotape as well as the photo , saying they were too unclear to identify Muhammad .", "predicted": -100, "true_label": "Defense lawyers had objected in earlier hearings to showing the videotape and the photo , saying they were too unclear to identify the person in them ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDefense lawyers had objected in earlier hearings to showing the videotape and the photo , saying they were too unclear to identify the person in them . ⇔ ", "response": "Defense lawyers had objected in earlier hearings , saying the tape was too unclear and the photo was too blurry to identify anyone .", "predicted": -100, "true_label": "Defense lawyers had objected in pretrial hearings to the videotape and the photo , saying they were too unclear to identify Muhammad ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of children in African clothing. ⇔ ", "response": "Children playing in a village in Africa.", "predicted": -100, "true_label": "A person in clothing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person in clothing ⇔ ", "response": "A person in a blue shirt and black pants", "predicted": -100, "true_label": "A group of children in African clothing."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAlthough it 's unclear whether Sobig was to blame , The New York Times also asked employees at its headquarters yesterday to shut down their computers because of \" system difficulties . \" ⇔ ", "response": "It was also unclear whether Sobig was to blame , but The New York Times asked employees at its headquarters to shut down their computers yesterday because of \" system difficulties . \"", "predicted": -100, "true_label": "The New York Times asked employees at its headquarters to shut down their computers yesterday because of \" computing system difficulties . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe New York Times asked employees at its headquarters to shut down their computers yesterday because of \" computing system difficulties . \" ⇔ ", "response": "As a result , employees were told to shut down their computers .", "predicted": -100, "true_label": "Although it 's unclear whether Sobig was to blame , The New York Times also asked employees at its headquarters yesterday to shut down their computers because of \" system difficulties . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有什么方法可以减少电脑辐射？ ⇔ ", "response": "电脑辐射有什么危害？", "predicted": -100, "true_label": "什么方法可以减少电脑辐射呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么方法可以减少电脑辐射呢 ⇔ ", "response": "减少电脑辐射的方法有哪些？", "predicted": -100, "true_label": "有什么方法可以减少电脑辐射？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么品牌的染发剂最好 ⇔ ", "response": "您使用过最好的染发剂是什么", "predicted": -100, "true_label": "什么牌子的染发剂最好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么牌子的染发剂最好 ⇔ ", "response": "进口染发剂什么牌子好", "predicted": -100, "true_label": "什么品牌的染发剂最好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFemale in a military uniform with her hair tied back. ⇔ ", "response": "a woman in a military uniform.", "predicted": -100, "true_label": "the female is wearing a ponytail"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nthe female is wearing a ponytail ⇔ ", "response": "The woman is wearing a hairpiece.", "predicted": -100, "true_label": "Female in a military uniform with her hair tied back."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n此图出自哪里 ⇔ ", "response": "图上画了谁", "predicted": -100, "true_label": "此图出自何处？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n此图出自何处？ ⇔ ", "response": "这幅图出自何处？", "predicted": -100, "true_label": "此图出自哪里"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该站于1903年7月1日在Stoneorlar至Glenties的多尼戈尔铁路线上开通。 ⇔ ", "response": "该站于1903年7月1日在Stoneorlar至Glenties的多尼戈尔铁路线上线。", "predicted": 1903, "true_label": "该站于1903年7月1日在多尼戈尔铁路公司的线路上开通，从Glenties到Stranorlar。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该站于1903年7月1日在多尼戈尔铁路公司的线路上开通，从Glenties到Stranorlar。 ⇔ ", "response": "该站于1903年7月1日在多尼戈尔铁路公司的线路上开通，从Glenties到Stranorlar。", "predicted": 1903, "true_label": "该站于1903年7月1日在Stoneorlar至Glenties的多尼戈尔铁路线上开通。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppo x ⇔ ", "response": "OPPO_X", "predicted": -100, "true_label": "oppofx"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppofx ⇔ ", "response": "OPPOFV", "predicted": -100, "true_label": "oppo x"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人把一种混合物倒进一个碗里，然后把它混合在一起。 ⇔ ", "response": "一个女人正在把食物混合物放到碗里。", "predicted": -100, "true_label": "一个女人正在煎锅里煮鸡蛋。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人正在煎锅里煮鸡蛋。 ⇔ ", "response": "一个女人在煎锅里煮鸡蛋。", "predicted": -100, "true_label": "一个女人把一种混合物倒进一个碗里，然后把它混合在一起。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman in red blouse is standing with small blond child in front of a small folding chalkboard. ⇔ ", "response": "A woman in a red blouse is standing near a child.", "predicted": -100, "true_label": "a woman stands with a child"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na woman stands with a child ⇔ ", "response": "A woman is carrying a child.", "predicted": -100, "true_label": "A woman in red blouse is standing with small blond child in front of a small folding chalkboard."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n找朋友借钱怎么开口说呢？ ⇔ ", "response": "怎么跟朋友说借这个钱", "predicted": -100, "true_label": "问朋友借钱好难，好难开口怎么办？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n问朋友借钱好难，好难开口怎么办？ ⇔ ", "response": "怎么办好难开口借朋友钱", "predicted": -100, "true_label": "找朋友借钱怎么开口说呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n零点四五乘四 ⇔ ", "response": "零点四五乘以四", "predicted": -100, "true_label": "零点四五乘零点二"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n零点四五乘零点二 ⇔ ", "response": "零点零零二点五乘零点二", "predicted": -100, "true_label": "零点四五乘四"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA crowd wearing orange cheering for their team in a stadium. ⇔ ", "response": "A group of people are gathered together.", "predicted": -100, "true_label": "People wearing orange."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople wearing orange. ⇔ ", "response": "A group of people in orange vests climb on top of each other near some trees.", "predicted": -100, "true_label": "A crowd wearing orange cheering for their team in a stadium."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is slicing bread ⇔ ", "response": "There is no man slicing bread", "predicted": -100, "true_label": "A person is slicing a piece of bread"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is slicing a piece of bread ⇔ ", "response": "There is no person slicing a piece of bread", "predicted": -100, "true_label": "A man is slicing bread"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo individuals use a photo kiosk ⇔ ", "response": "A man uses a photo kiosk.", "predicted": -100, "true_label": "There are two individuals"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are two individuals ⇔ ", "response": "Two individuals are standing outside on a rainy day", "predicted": -100, "true_label": "Two individuals use a photo kiosk"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGOODGOOD ⇔ ", "response": "Goodday", "predicted": -100, "true_label": "Thatsgood"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThatsgood ⇔ ", "response": "THATGIDDOUTBIGGER", "predicted": -100, "true_label": "GOODGOOD"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHIBABY ⇔ ", "response": "BYHIBABY", "predicted": -100, "true_label": "HIKING"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHIKING ⇔ ", "response": "GOING", "predicted": -100, "true_label": "HIBABY"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMayor Joe T. Parker said late Thursday that the three workers were two men and a woman who were inside the building when the first blast occurred . ⇔ ", "response": "Three city workers were injured when a second blast struck the building early Thursday , Mayor Joe T. Parker said . The three workers were inside the building when the first blast occurred .", "predicted": -100, "true_label": "The missing workers , two men and a woman , were inside the building when the first blast occurred , Mayor Joe T. Parker said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe missing workers , two men and a woman , were inside the building when the first blast occurred , Mayor Joe T. Parker said . ⇔ ", "response": "The woman and man were inside the building when the first blast occurred , Mayor Joe T. Parker said .", "predicted": -100, "true_label": "Mayor Joe T. Parker said late Thursday that the three workers were two men and a woman who were inside the building when the first blast occurred ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of people are walking by a van. ⇔ ", "response": "A group of people walk past a van with the engine still running.", "predicted": -100, "true_label": "Some people outdoors."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome people outdoors. ⇔ ", "response": "Several people are playing in the water.", "predicted": -100, "true_label": "A group of people are walking by a van."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n37分钟后叫我 ⇔ ", "response": "9:37叫我说night", "predicted": 9, "true_label": "三十分钟计时开始"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n三十分钟计时开始 ⇔ ", "response": "暂停计时器", "predicted": -100, "true_label": "37分钟后叫我"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n印度的建筑倒塌；25人害怕被困。 ⇔ ", "response": "德里的一栋楼倒塌，25人在废墟中。", "predicted": 25, "true_label": "印度建筑倒塌造成14人死亡，数十人被困。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n印度建筑倒塌造成14人死亡，数十人被困。 ⇔ ", "response": "印度工人在一个建筑项目中工作。", "predicted": -100, "true_label": "印度的建筑倒塌；25人害怕被困。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n发短信给妈 ⇔ ", "response": "打电话给妈", "predicted": -100, "true_label": "搜索信息"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n搜索信息 ⇔ ", "response": "搜索", "predicted": -100, "true_label": "发短信给妈"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA crowd wearing orange cheering for their team in a stadium. ⇔ ", "response": "A group of people are gathered together.", "predicted": -100, "true_label": "Fans cheering on their team at the big game."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFans cheering on their team at the big game. ⇔ ", "response": "People are watching a football game.", "predicted": -100, "true_label": "A crowd wearing orange cheering for their team in a stadium."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is applying makeup to her face ⇔ ", "response": "There is no girl applying makeup to her face", "predicted": -100, "true_label": "There is no girl applying makeup to her face"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no girl applying makeup to her face ⇔ ", "response": "A girl is applying makeup to her face", "predicted": -100, "true_label": "A girl is applying makeup to her face"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGreen tea consumption is associated with decreased risk of breast, pancreatic, colon, oesophageal, and lung cancers in humans. ⇔ ", "response": "Green tea consumption reduces the risk of lung cancer.", "predicted": -100, "true_label": "Tea protects from some diseases."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTea protects from some diseases. ⇔ ", "response": "Scientists have discovered that drinking tea protects against heart disease by improving the function of the artery walls.", "predicted": -100, "true_label": "Green tea consumption is associated with decreased risk of breast, pancreatic, colon, oesophageal, and lung cancers in humans."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is quickly chopping some mushrooms with a knife ⇔ ", "response": "A man is slowly chopping some mushrooms with a knife", "predicted": -100, "true_label": "A person is quickly chopping mushrooms with a knife"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is quickly chopping mushrooms with a knife ⇔ ", "response": "A person is chopping mushrooms with a knife", "predicted": -100, "true_label": "A man is quickly chopping some mushrooms with a knife"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n二十年前 ⇔ ", "response": "明天上午八点", "predicted": -100, "true_label": "明月何时照我还"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明月何时照我还 ⇔ ", "response": "明月去年照我还", "predicted": -100, "true_label": "二十年前"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe victims were not identified , although authorities said at least one of the bodies had been there about a year . ⇔ ", "response": "The victims were not identified , although authorities said you 'll find out more about them at 11 a.m. : weeknights : station : KGO-TV channel 9 . 11 : hour : 11 : minute : 20 : location : Mill Valley .", "predicted": 11, "true_label": "The two victims buried in the yard were not identified , though authorities said at least one has been there about a year ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe two victims buried in the yard were not identified , though authorities said at least one has been there about a year . ⇔ ", "response": "The victims were not identified, but a police official said one had been there about a year .", "predicted": -100, "true_label": "The victims were not identified , although authorities said at least one of the bodies had been there about a year ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWe feel so strongly about this issue that we are suspending sales and distribution of SCOLinux until these issues are resolved , \" Sontag said . ⇔ ", "response": "\" We feel strongly about this issue and until they are resolved , we are suspending sales and distribution of SCOLinux , \" Sontag said .", "predicted": -100, "true_label": "We feel so strongly about this issue that we are suspending sales and distribution of SCO Linux until these issues are resolved . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWe feel so strongly about this issue that we are suspending sales and distribution of SCO Linux until these issues are resolved . \" ⇔ ", "response": "\" Sco says it will suspend sales and distribution of sco linux until the issues are resolved , \" the company added .", "predicted": -100, "true_label": "We feel so strongly about this issue that we are suspending sales and distribution of SCOLinux until these issues are resolved , \" Sontag said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n小游戏app上找不到玩过的游戏 ⇔ ", "response": "微信小游戏封号潮持续！又有一款游戏被关入小黑屋！", "predicted": -100, "true_label": "脚上有伤能拍上去几天就好的软件"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n脚上有伤能拍上去几天就好的软件 ⇔ ", "response": "能拍上去几天就好的软件", "predicted": -100, "true_label": "小游戏app上找不到玩过的游戏"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你是人吗 ⇔ ", "response": "你是人啊", "predicted": -100, "true_label": "这是什么东西"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这是什么东西 ⇔ ", "response": "你喜欢我吗东西", "predicted": -100, "true_label": "你是人吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天你吃了晚饭吗 ⇔ ", "response": "今天你吃晚饭吗", "predicted": -100, "true_label": "今天天气很晚饭吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气很晚饭吗 ⇔ ", "response": "今天天气很好吗", "predicted": -100, "true_label": "今天你吃了晚饭吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is riding a horse in the desert ⇔ ", "response": "A man is riding a horse", "predicted": -100, "true_label": "A horse is being ridden by a guy"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA horse is being ridden by a guy ⇔ ", "response": "A horse is being tamed by a guy", "predicted": -100, "true_label": "A man is riding a horse in the desert"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA high school band getting ready to perform. ⇔ ", "response": "A band is playing", "predicted": -100, "true_label": "A high school band prepares to play in front of a large crowd."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA high school band prepares to play in front of a large crowd. ⇔ ", "response": "A band plays at a football game.", "predicted": -100, "true_label": "A high school band getting ready to perform."}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor9splusota ⇔ ", "response": "OPPO r9s plus", "predicted": 9, "true_label": "￥lsquo"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n￥lsquo ⇔ ", "response": "rsquo", "predicted": -100, "true_label": "oppor9splusota"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are eating ⇔ ", "response": "A man and a woman are dining", "predicted": -100, "true_label": "A man and a woman are not eating"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are not eating ⇔ ", "response": "A man and a woman are eating", "predicted": -100, "true_label": "A man and a woman are eating"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它沿着旧的Sohmer钢琴厂，在核桃街下，沿着主街。 ⇔ ", "response": "它沿着旧的Sohmer钢琴厂下去，位于核桃街和主要街道之间。", "predicted": -100, "true_label": "它沿着老Sommer钢琴厂，在核桃街和主街沿线。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它沿着老Sommer钢琴厂，在核桃街和主街沿线。 ⇔ ", "response": "它沿着旧sommer钢琴厂沿线，在核桃街和主街沿线。", "predicted": -100, "true_label": "它沿着旧的Sohmer钢琴厂，在核桃街下，沿着主街。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去学校 ⇔ ", "response": "我今天去学校了了", "predicted": -100, "true_label": "我明天去学孝"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我明天去学孝 ⇔ ", "response": "明天去学车", "predicted": -100, "true_label": "我今天去学校"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜欢吃苹果 ⇔ ", "response": "今天的天气真好", "predicted": -100, "true_label": "我喜焑吃苹果"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜焑吃苹果 ⇔ ", "response": "我喜欢吃苹果", "predicted": -100, "true_label": "我喜欢吃苹果"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe most common form of estrogen given to postmenopausal women is conjugated estrogen, which is made from the urine of pregnant mares, and the most popular brand of conjugated estrogen is Premarin. ⇔ ", "response": "Estrogen is given to postmenopausal women.", "predicted": -100, "true_label": "Estrogen is used by postmenopausal women."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nEstrogen is used by postmenopausal women. ⇔ ", "response": "Most notably, the Women's Health Initiative recently found that postmenopausal women taking estrogen with progestin have an increased risk of breast cancer as well as heart attack, stroke, and blood clots.", "predicted": -100, "true_label": "The most common form of estrogen given to postmenopausal women is conjugated estrogen, which is made from the urine of pregnant mares, and the most popular brand of conjugated estrogen is Premarin."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no person boiling noodles ⇔ ", "response": "A person is boiling noodles", "predicted": -100, "true_label": "A woman is boiling noodles in water"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is boiling noodles in water ⇔ ", "response": "A person is boiling noodles", "predicted": -100, "true_label": "There is no person boiling noodles"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n谁有铠甲勇士之帝皇侠电影版 ⇔ ", "response": "铠甲勇士帝皇龙", "predicted": -100, "true_label": "谁有铠甲勇士之帝皇侠电影，不要花絮。电影"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n谁有铠甲勇士之帝皇侠电影，不要花絮。电影 ⇔ ", "response": "铠甲勇士帝皇侠电影在哪里看", "predicted": -100, "true_label": "谁有铠甲勇士之帝皇侠电影版"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe government and private technology experts warned Wednesday that hackers plan to attack thousands of Web sites on Sunday in a loosely coordinated \" contest \" that could disrupt Internet traffic . ⇔ ", "response": "The government and private technology experts said that hackers plan to attack thousands of sites on Sunday in a loosely coordinated \"contest, \" which could disrupt Internet traffic .", "predicted": -100, "true_label": "THE US government and private technology experts have warned that hackers plan to attack thousands of websites on Sunday in a loosely co-ordinated \" contest \" that could disrupt Internet traffic ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTHE US government and private technology experts have warned that hackers plan to attack thousands of websites on Sunday in a loosely co-ordinated \" contest \" that could disrupt Internet traffic . ⇔ ", "response": "A government and private technology expert said hackers were planning to attack thousands of websites around the world on Sunday in a loosely co-ordinated \" contest \" that could disrupt Internet traffic .", "predicted": -100, "true_label": "The government and private technology experts warned Wednesday that hackers plan to attack thousands of Web sites on Sunday in a loosely coordinated \" contest \" that could disrupt Internet traffic ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man rides a black motorcycle while wearing a helmet and a van rides beside him in the distance. ⇔ ", "response": "A man rides a motorcycle.", "predicted": -100, "true_label": "A motorcycle and a van ride next to each other."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA motorcycle and a van ride next to each other. ⇔ ", "response": "A motorcycle is parked outside of a van.", "predicted": -100, "true_label": "A man rides a black motorcycle while wearing a helmet and a van rides beside him in the distance."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n七龙珠小说 ⇔ ", "response": "七龙珠息小说", "predicted": -100, "true_label": "月亮抱着太阳小说中文版"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n月亮抱着太阳小说中文版 ⇔ ", "response": "月亮抱着太阳", "predicted": -100, "true_label": "七龙珠小说"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe flagship \"Atocha\" and eight other ships loaded with treasures went down, together with the crews of 550. ⇔ ", "response": "More than 500 people went down with the \"Atocha\" in 1787.", "predicted": 500, "true_label": "People were killed in a collision between ships."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople were killed in a collision between ships. ⇔ ", "response": "The al-Salam 95 had been waiting to enter the port, following its journey from Jeddah, when the two ships collided.", "predicted": 95, "true_label": "The flagship \"Atocha\" and eight other ships loaded with treasures went down, together with the crews of 550."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhen an earthquake rumbled off the coast of Hokkaido in Japan in July of 1993, the resulting tsunami hit just three to five minutes later, killing 202 people who were trying to flee for higher ground. ⇔ ", "response": "A tsunami hit Hokkaido in Japan.", "predicted": -100, "true_label": "An earthquake occurred on the coast of Hokkaido, Japan."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn earthquake occurred on the coast of Hokkaido, Japan. ⇔ ", "response": "A great earthquake occurred at 19:50:08 (UTC) on Wednesday, September 25, 2003. The magnitude 8.0 event has been located in the Sea of Japan, near the coast of Hokkaido, Japan.", "predicted": 19, "true_label": "When an earthquake rumbled off the coast of Hokkaido in Japan in July of 1993, the resulting tsunami hit just three to five minutes later, killing 202 people who were trying to flee for higher ground."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n台湾新闻 ⇔ ", "response": "台湾新闻界", "predicted": -100, "true_label": "美国选举新闻"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n美国选举新闻 ⇔ ", "response": "美国选举报道", "predicted": -100, "true_label": "台湾新闻"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nInception是一部2010年的科幻电影，由艾玛托马斯联合制作，由克里斯托弗诺兰编剧，联合制作和导演。 ⇔ ", "response": "Inception是一部2010年的电影，由艾玛托马斯联合制作，由克里斯托弗诺兰编剧，联合制作和导演。", "predicted": 2010, "true_label": "Inception是一部2010年的科幻电影，由Emma Thomas联合制作，由Christopher Nolan编写，共同制作并上演。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nInception是一部2010年的科幻电影，由Emma Thomas联合制作，由Christopher Nolan编写，共同制作并上演。 ⇔ ", "response": "Inception是一部电影，由Emma Thomas联合制作，Christopher Nolan编写，并首次在2010年上映。", "predicted": 2010, "true_label": "Inception是一部2010年的科幻电影，由艾玛托马斯联合制作，由克里斯托弗诺兰编剧，联合制作和导演。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在2015年的地区选举中，在托西被联邦党排除在外，吉多尼被选入贝卢诺省的威尼托区域委员会。 ⇔ ", "response": "在2015年的区域选举中，吉多尼在联邦党被排除在托西之外后当选为贝卢诺威尼托区域委员会的成员。", "predicted": 2015, "true_label": "在2015年的地区选举中，在托西被联邦政党搁置后，吉多尼当选为贝卢诺省的威尼托区域委员会成员。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在2015年的地区选举中，在托西被联邦政党搁置后，吉多尼当选为贝卢诺省的威尼托区域委员会成员。 ⇔ ", "response": "在2015年的区域选举中，当托西在联邦政党中占据一席之地时，吉多尼当选为贝卢诺省的威尼托区域委员会成员。", "predicted": 2015, "true_label": "在2015年的地区选举中，在托西被联邦党排除在外，吉多尼被选入贝卢诺省的威尼托区域委员会。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气不错 ⇔ ", "response": "今天天气不做", "predicted": -100, "true_label": "今天天气不做"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气不做 ⇔ ", "response": "今天天气不错", "predicted": -100, "true_label": "今天天气不错"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSpecifically, the October 4, 1994 earthquake (as the October 12, 1963 earthquake) ocurred at the Pacific side boundary of the Kuril arc (the South Kuril Islands) of the Okhotsk plate. ⇔ ", "response": "An earthquake occurred on the ocean floor.", "predicted": -100, "true_label": "An earthquake occurred in 1994."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn earthquake occurred in 1994. ⇔ ", "response": "On January 18, 1994, a magnitude 7.1 earthquake struck the Southern California coast.", "predicted": 18, "true_label": "Specifically, the October 4, 1994 earthquake (as the October 12, 1963 earthquake) ocurred at the Pacific side boundary of the Kuril arc (the South Kuril Islands) of the Okhotsk plate."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n笑话我的 ⇔ ", "response": "笑话您的吧", "predicted": -100, "true_label": "给我讲个笑话好不好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n给我讲个笑话好不好 ⇔ ", "response": "笑话有什么好讲的", "predicted": -100, "true_label": "笑话我的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我们来玩成语接龙 ⇔ ", "response": "我们来玩成语了", "predicted": -100, "true_label": "成语接龙大全"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n成语接龙大全 ⇔ ", "response": "成语大全", "predicted": -100, "true_label": "我们来玩成语接龙"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我这配置能玩洛奇英雄传吗？ ⇔ ", "response": "这配置能玩洛奇英雄传吗？", "predicted": -100, "true_label": "我这配置可以玩洛奇英雄传吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我这配置可以玩洛奇英雄传吗 ⇔ ", "response": "我这配置可以运行洛奇英雄传吗", "predicted": -100, "true_label": "我这配置能玩洛奇英雄传吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is not walking the dog ⇔ ", "response": "A woman is walking the dog", "predicted": -100, "true_label": "The woman is walking the dog"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is walking the dog ⇔ ", "response": "A woman is walking the dog down the street", "predicted": -100, "true_label": "The woman is not walking the dog"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃兔肉吗 ⇔ ", "response": "怀孕能吃兔胎吗", "predicted": -100, "true_label": "怀孕能吃香菜吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃香菜吗 ⇔ ", "response": "怀孕能吃草莓吗", "predicted": -100, "true_label": "怀孕能吃兔肉吗"}

eval_gen/gpu_results/results_gpu_4.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n吃柿子有什么好处 ⇔ ", "response": "吃柿子坏处是什么", "predicted": -100, "true_label": "怀孕能吃小龙虾吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃小龙虾吗 ⇔ ", "response": "小龙虾能吃吗", "predicted": -100, "true_label": "吃柿子有什么好处"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n又向自己的脸颊和鼻子使劲打了五六拳，打得血肉模糊。 ⇔ ", "response": "接着，他又向自己的脸颊和鼻子使劲打了五六拳，打了没有红，打了却紫，打得血肉模糊。", "predicted": -100, "true_label": "于是他两手去抓自己的胡子，把胡子揪掉了一半，然后又向自己的面颊和鼻子一连打了五六拳，打得自己满脸是血。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n于是他两手去抓自己的胡子，把胡子揪掉了一半，然后又向自己的面颊和鼻子一连打了五六拳，打得自己满脸是血。 ⇔ ", "response": "于是他向自己的脸打，打了五六拳，打得自己满脸是血。", "predicted": -100, "true_label": "又向自己的脸颊和鼻子使劲打了五六拳，打得血肉模糊。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n德国一季度实现增长，躲过经济衰退 ⇔ ", "response": "德国经济衰退避免", "predicted": -100, "true_label": "多亏了德国，欧元区避免了第一季度的衰退。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n多亏了德国，欧元区避免了第一季度的衰退。 ⇔ ", "response": "否则，欧元区将经历负增长。", "predicted": -100, "true_label": "德国一季度实现增长，躲过经济衰退"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样炒青菜才好吃啊？ ⇔ ", "response": "怎样炒菜心才好吃？", "predicted": -100, "true_label": "青菜怎么炒才好吃?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n青菜怎么炒才好吃? ⇔ ", "response": "你怎么炒青菜好吃？", "predicted": -100, "true_label": "怎样炒青菜才好吃啊？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhen it was revealed that under her government, irregular special forces, possibly including the Hezbollah, were illegally financed to fight the PKK, Ciller reacted by saying she had done no wrong and would do the same again. ⇔ ", "response": "Hezbollah operates in Turkey.", "predicted": -100, "true_label": "A French political party was illegally financed."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA French political party was illegally financed. ⇔ ", "response": "The money came from Xenophon Tukyulagi, a French lawstudent who has become embroiled with several Frenchpolitical parties as an activist.", "predicted": -100, "true_label": "When it was revealed that under her government, irregular special forces, possibly including the Hezbollah, were illegally financed to fight the PKK, Ciller reacted by saying she had done no wrong and would do the same again."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n圆脸小眼睛女孩适合什么发型 ⇔ ", "response": "女孩适合什么发型", "predicted": -100, "true_label": "圆脸小眼睛不高适合什么发型和刘海。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n圆脸小眼睛不高适合什么发型和刘海。 ⇔ ", "response": "圆脸小眼睛适合什么发型和刘海。", "predicted": -100, "true_label": "圆脸小眼睛女孩适合什么发型"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n英国马恩岛航空公司的历史可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。 ⇔ ", "response": "英国马恩岛航空公司的历史可以追溯到1991年3月，当时马恩岛航空公司创建了EICA，以便在英国境内扩展其航线并飞行。", "predicted": 1991, "true_label": "马恩岛航空公司可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n马恩岛航空公司可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。 ⇔ ", "response": "1991年3月，马恩岛航空公司创建了一家区域航空公司，以便在英国境内扩展和飞行航线。", "predicted": 1991, "true_label": "英国马恩岛航空公司的历史可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n美女主播写真集 ⇔ ", "response": "美女主播��真集", "predicted": -100, "true_label": "赛文奥特曼的照片"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n赛文奥特曼的照片 ⇔ ", "response": "宇宙怪兽祖尔格龙的照片", "predicted": -100, "true_label": "美女主播写真集"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n娜塔莎再也不跟索尼娅说话，躲着她。 ⇔ ", "response": "也许娜塔莎会改变主意，跟索尼娅说话，躲着她不是主要的原因。", "predicted": -100, "true_label": "娜塔莎不再跟索尼娅讲话了，并且老躲着她。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n娜塔莎不再跟索尼娅讲话了，并且老躲着她。 ⇔ ", "response": "娜塔莎也不再跟索尼娅讲话了，并且老躲着她。", "predicted": -100, "true_label": "娜塔莎再也不跟索尼娅说话，躲着她。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打开什么值得买里签到好不好 ⇔ ", "response": "什么值得买里签到的好物好不好", "predicted": -100, "true_label": "那你在干嘛"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n那你在干嘛 ⇔ ", "response": "你在干嘛呢", "predicted": -100, "true_label": "打开什么值得买里签到好不好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n阿富汗首都自杀式炸弹袭击造成6人死亡 ⇔ ", "response": "阿富汗首都爆炸造成至少16人死亡", "predicted": 16, "true_label": "自杀式袭击者在阿富汗清真寺炸死37人"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n自杀式袭击者在阿富汗清真寺炸死37人 ⇔ ", "response": "自杀式袭击者在阿富汗教堂炸死37人", "predicted": 37, "true_label": "阿富汗首都自杀式炸弹袭击造成6人死亡"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n刻薄的反义词是什么 ⇔ ", "response": "意思是是什么", "predicted": -100, "true_label": "刻薄的反义词是什么？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n刻薄的反义词是什么？ ⇔ ", "response": "什么是不刻薄", "predicted": -100, "true_label": "刻薄的反义词是什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n送前男友什么生日礼物好 ⇔ ", "response": "生日送什么礼物好", "predicted": -100, "true_label": "送什么生日礼物给前男友好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n送什么生日礼物给前男友好 ⇔ ", "response": "前男友过生日送什么礼物", "predicted": -100, "true_label": "送前男友什么生日礼物好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe former president also gave numerous speeches in 2002 without compensation , said his spokesman Jim Kennedy . ⇔ ", "response": "In 2002 , Bush gave dozens of paid speeches around the country , his spokesman said .", "predicted": 2002, "true_label": "His spokesman Jim Kennedy said the former president also gave more than 70 speeches in 2002 without compensation ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHis spokesman Jim Kennedy said the former president also gave more than 70 speeches in 2002 without compensation . ⇔ ", "response": "Kennedy said that in addition to the speech at his alma mater , Bush also gave more than 70 paid speeches in 2002 .", "predicted": 70, "true_label": "The former president also gave numerous speeches in 2002 without compensation , said his spokesman Jim Kennedy ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhen replacing a single tooth, an implant has an advantage over fixed bridgework in that healthy teeth adjacent to the missing tooth do not have to be prepped (drilled down). ⇔ ", "response": "Implants are used to replace single teeth.", "predicted": -100, "true_label": "Tooth implants have advantages."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTooth implants have advantages. ⇔ ", "response": "Although there are risks involved with tooth implants, such as infection, bad fit or fit loss, and anesthesia reactions, the advantage of tooth implants over other options outweighs the disadvantages, the authors write.", "predicted": -100, "true_label": "When replacing a single tooth, an implant has an advantage over fixed bridgework in that healthy teeth adjacent to the missing tooth do not have to be prepped (drilled down)."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAMSORRY ⇔ ", "response": "Imsorry", "predicted": -100, "true_label": "IMSORRY"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIMSORRY ⇔ ", "response": "SORRY", "predicted": -100, "true_label": "AMSORRY"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n界上最小的国家 ⇔ ", "response": "中国最北端的第一家私人旅行社", "predicted": -100, "true_label": "���上有鬼吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n界上有鬼吗 ⇔ ", "response": "有鬼吗", "predicted": -100, "true_label": "界上最小的国家"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is singing and playing a guitar ⇔ ", "response": "There is no man singing and playing a guitar", "predicted": -100, "true_label": "A man is playing a guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is playing a guitar ⇔ ", "response": "There is no man playing a guitar", "predicted": -100, "true_label": "A man is singing and playing a guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFemale gymnasts warm up before a competition. ⇔ ", "response": "The women are doing gymnastics", "predicted": -100, "true_label": "Gymnasts get ready for a competition."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGymnasts get ready for a competition. ⇔ ", "response": "Two female artistic gymnasts doing warm up exercises.", "predicted": -100, "true_label": "Female gymnasts warm up before a competition."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天天酷跑百变萝莉和超能少年哪个好 ⇔ ", "response": "天天酷跑哪个角色好", "predicted": -100, "true_label": "天天酷跑超能少年和百变萝莉哪个更好一点"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天天酷跑超能少年和百变萝莉哪个更好一点 ⇔ ", "response": "天天酷跑超能少年和百变萝莉哪个好", "predicted": -100, "true_label": "天天酷跑百变萝莉和超能少年哪个好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her. ⇔ ", "response": "A girl is smiling.", "predicted": -100, "true_label": "The young girl is wearing a red shirt."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe young girl is wearing a red shirt. ⇔ ", "response": "A young girl with a red shirt and black pants is walking behind a boy in blue jeans and a boy in a red and white striped shirt.", "predicted": -100, "true_label": "A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGateway will release new Profile 4 systems with the new Intel technology on Wednesday . ⇔ ", "response": "Gateway will begin selling the new systems Wednesday .", "predicted": -100, "true_label": "Gateway 's all-in-one PC , the Profile 4 , also now features the new Intel technology ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGateway 's all-in-one PC , the Profile 4 , also now features the new Intel technology . ⇔ ", "response": "The Profile 4 is Gateway 's first all-in-one PC to feature Intel 's technology .", "predicted": 4, "true_label": "Gateway will release new Profile 4 systems with the new Intel technology on Wednesday ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样举报拼多多客服 ⇔ ", "response": "怎么投诉拼多多的客服", "predicted": -100, "true_label": "软件太多费流量怎么办"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n软件太多费流量怎么办 ⇔ ", "response": "软件安装费流量怎么办", "predicted": -100, "true_label": "怎样举报拼多多客服"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBogdanowice是波兰的一个村庄，位于Głubczyce县（CommuneGłubczyce）的Opole Voivodeship。 ⇔ ", "response": "Bogdanowice是波兰的一个村庄，位于Głubczyce县的Opole Voivodeship的县。", "predicted": -100, "true_label": "Bogdanowice是一个位于波兰Głubczyce县的村庄，位于Opole Voivodeship（GminaGłubczyce）。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBogdanowice是一个位于波兰Głubczyce县的村庄，位于Opole Voivodeship（GminaGłubczyce）。 ⇔ ", "response": "Bogdanowice是一个位于波兰Głubczyce县的村庄，位于Opole Voivodeship（GminaGłubczyce）。", "predicted": -100, "true_label": "Bogdanowice是波兰的一个村庄，位于Głubczyce县（CommuneGłubczyce）的Opole Voivodeship。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n九千公顷等于多少平方千米 ⇔ ", "response": "一千公顷等于多少平方千米", "predicted": -100, "true_label": "分钟等于多少秒"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n分钟等于多少秒 ⇔ ", "response": "小时等于多少分钟", "predicted": -100, "true_label": "九千公顷等于多少平方千米"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n巴基斯坦汽车炸弹炸死17人 ⇔ ", "response": "汽车炸弹炸死巴基斯坦���备部队16人", "predicted": 16, "true_label": "巴基斯坦自杀式袭击者炸死10人"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n巴基斯坦自杀式袭击者炸死10人 ⇔ ", "response": "巴基斯坦袭击造成至少13人死亡", "predicted": 13, "true_label": "巴基斯坦汽车炸弹炸死17人"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree men are dancing ⇔ ", "response": "Three men are dancing in front of a television set", "predicted": -100, "true_label": "The men are not dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe men are not dancing ⇔ ", "response": "Some men are dancing gracefully to the music", "predicted": -100, "true_label": "Three men are dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is stirring meat in a bowl ⇔ ", "response": "The woman is not stirring meat in a bowl", "predicted": -100, "true_label": "There is no lady mixing a meat mixture in a bowl"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no lady mixing a meat mixture in a bowl ⇔ ", "response": "A lady is mixing a meat mixture in a bowl", "predicted": -100, "true_label": "The woman is stirring meat in a bowl"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nZotov said that Russia would agree to the complete withdrawal of its troops from Latvia by August 31, 1994 if Latvia allowed Russia to use the Skrunda radar for the next four years and dismantle it within the following 18 months. ⇔ ", "response": "Zotov agreed to let Latvia keep the Skrunda radar.", "predicted": -100, "true_label": "Russian troops withdrew from Latvia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRussian troops withdrew from Latvia. ⇔ ", "response": "Latvia grudgingly agreed to let Russia keep the Skrunda radar; the Russians switched it off in 1998, then had 18 months to dismantle it.", "predicted": 1998, "true_label": "Zotov said that Russia would agree to the complete withdrawal of its troops from Latvia by August 31, 1994 if Latvia allowed Russia to use the Skrunda radar for the next four years and dismantle it within the following 18 months."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPO r11s plus ⇔ ", "response": "OPPOR11sPlus", "predicted": 11, "true_label": "oppor11plus"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor11plus ⇔ ", "response": "OPPO r11 plus", "predicted": 11, "true_label": "OPPO r11s plus"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMan and woman rowing on surfboards third rower in distance. ⇔ ", "response": "A man and woman are rowing their surfboard boats.", "predicted": -100, "true_label": "Couples rowing on surfboards."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCouples rowing on surfboards. ⇔ ", "response": "Four couples are rowing.", "predicted": -100, "true_label": "Man and woman rowing on surfboards third rower in distance."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is fixing a silencer to a gun ⇔ ", "response": "A man is attaching a gun to a silencer", "predicted": -100, "true_label": "A man is applying a silencer to a gun"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is applying a silencer to a gun ⇔ ", "response": "A man is completing a gun", "predicted": -100, "true_label": "A man is fixing a silencer to a gun"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n教堂边门口，他立住脚，发现我上气不接下气。 ⇔ ", "response": "教堂门口，他停住脚步，看到我两口气出不完。", "predicted": -100, "true_label": "在教堂墓地的边门口，他停了下来；他发现我已经完全上气不接下气了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在教堂墓地的边门口，他停了下来；他发现我已经完全上气不接下气了。 ⇔ ", "response": "他和我并排靠在教堂墓地的边门口；他发现我已经完全气喘吁吁了。", "predicted": -100, "true_label": "教堂边门口，他立住脚，发现我上气不接下气。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no girl playing the piano ⇔ ", "response": "A girl is playing the piano", "predicted": -100, "true_label": "A girl is playing the piano"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA girl is playing the piano ⇔ ", "response": "A little girl is playing a grand piano on stage", "predicted": -100, "true_label": "There is no girl playing the piano"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA kitten is playing with a blue rope toy ⇔ ", "response": "There is no kitten playing with a blue rope toy", "predicted": -100, "true_label": "The kitten is not playing with a toy"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe kitten is not playing with a toy ⇔ ", "response": "The kitten is playing with a toy", "predicted": -100, "true_label": "A kitten is playing with a blue rope toy"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo men are fighting in a cow pasture ⇔ ", "response": "Two men are fighting in a cow pasture and wearing hats", "predicted": -100, "true_label": "Nobody is fighting in a cow pasture"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNobody is fighting in a cow pasture ⇔ ", "response": "A bunch of people are fighting in a cow pasture", "predicted": -100, "true_label": "Two men are fighting in a cow pasture"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is boiling noodles ⇔ ", "response": "A woman is boiling noodles in water", "predicted": -100, "true_label": "There is no person boiling noodles"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no person boiling noodles ⇔ ", "response": "A person is boiling noodles", "predicted": -100, "true_label": "A person is boiling noodles"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA roofer in a gray sweatshirt and orange hat walks on a unfinished roof at a lake-side home. ⇔ ", "response": "A man repairs a roof.", "predicted": -100, "true_label": "The roofer is outside working."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe roofer is outside working. ⇔ ", "response": "An adult man wearing a green hat and tight shirt looks over his shoulder while standing outside in front of a red house.", "predicted": -100, "true_label": "A roofer in a gray sweatshirt and orange hat walks on a unfinished roof at a lake-side home."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nItaly's Alberto Tomba pulled out of the 2nd race. ⇔ ", "response": "Tomba pulled out of the second race.", "predicted": -100, "true_label": "Alberto Tomba won a ski race."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAlberto Tomba won a ski race. ⇔ ", "response": "Miller has redefined alpine skiing with his aggressive style, the likes of which hasn't been seen since Alberto Tomba first began to dominate slalom and giant slalom in the late eighties and throughout the early nineties, on route to 50 World Cup wins in those two disciplines.", "predicted": 50, "true_label": "Italy's Alberto Tomba pulled out of the 2nd race."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOZMXP ⇔ ", "response": "oppo zx", "predicted": -100, "true_label": "oppo x"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppo x ⇔ ", "response": "OPPO_X", "predicted": -100, "true_label": "OPPOZMXP"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is cutting a box ⇔ ", "response": "A man is cutting a shape out of wood", "predicted": -100, "true_label": "There is no man cutting a box"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man cutting a box ⇔ ", "response": "A man is cutting a box apart", "predicted": -100, "true_label": "A man is cutting a box"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nareyouok ⇔ ", "response": "areyou", "predicted": -100, "true_label": "whatdoyoudo"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nwhatdoyoudo ⇔ ", "response": "WHATDOYOUDO", "predicted": -100, "true_label": "areyouok"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n去哪个网站可以免费下载歌曲 ⇔ ", "response": "免费歌曲下载网站哪个最好", "predicted": -100, "true_label": "哪个网站可以免费下载完整版歌曲"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n哪个网站可以免费下载完整版歌曲 ⇔ ", "response": "哪里可以免费下载电影和歌曲", "predicted": -100, "true_label": "去哪个网站可以免费下载歌曲"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公园散步了。 ⇔ ", "response": "我今天去公园散b步了。", "predicted": -100, "true_label": "我今天去公元散步了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公元散步了。 ⇔ ", "response": "我今天去公园散步了。", "predicted": -100, "true_label": "我今天去公园散步了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一只湿漉漉的金发狗在岸边举着一根棍子。 ⇔ ", "response": "一只狗站在水中，举着一根棍子。", "predicted": -100, "true_label": "这只棕色湿漉漉的狗嘴里叼着一根棍子，正在水旁的沙子里奔跑。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这只棕色湿漉漉的狗嘴里叼着一根棍子，正在水旁的沙子里奔跑。 ⇔ ", "response": "一只棕色、湿漉漉的狗嘴里叼着一根棍子，在沙子和水之间奔跑。", "predicted": -100, "true_label": "一只湿漉漉的金发狗在岸边举着一根棍子。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree women are sitting on a green bench looking out at the coast. ⇔ ", "response": "The women are at the coast.", "predicted": -100, "true_label": "The women are outdoors."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe women are outdoors. ⇔ ", "response": "The women are outside on a beautiful day.", "predicted": -100, "true_label": "Three women are sitting on a green bench looking out at the coast."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n几点了呀 ⇔ ", "response": "几点了呀呀呀呀呀呀呀", "predicted": -100, "true_label": "感恩节是几月几号"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n感恩节是几月几号 ⇔ ", "response": "感恩节是什么节日", "predicted": -100, "true_label": "几点了呀"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA mountainous photo is complete with a blue sky. ⇔ ", "response": "A mountain range with a blue sky in the distance.", "predicted": -100, "true_label": "Someone has taken a photo."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone has taken a photo. ⇔ ", "response": "A person holding an electronic device in front of an electronic device.", "predicted": -100, "true_label": "A mountainous photo is complete with a blue sky."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nElderly woman in blue apron balances a basket on her head on a sidewalk while talking to a woman dressed in black. ⇔ ", "response": "The woman is balancing a basket on her head.", "predicted": -100, "true_label": "Elderly woman is balancing something on her head while having a conversation."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nElderly woman is balancing something on her head while having a conversation. ⇔ ", "response": "A woman with an elderly mother is balancing a bag on her head while talking to her.", "predicted": -100, "true_label": "Elderly woman in blue apron balances a basket on her head on a sidewalk while talking to a woman dressed in black."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n十一月四日风雨大作 ⇔ ", "response": "十一月四日风雨小", "predicted": -100, "true_label": "长风破浪会有时下一句"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n长风破浪会有时下一句 ⇔ ", "response": "功夫不负有心人", "predicted": -100, "true_label": "十一月四日风雨大作"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n练瑜伽对女人有什么好处？ ⇔ ", "response": "女人做什么运动好", "predicted": -100, "true_label": "练瑜伽对女人有什么好处?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n练瑜伽对女人有什么好处? ⇔ ", "response": "练习瑜伽对腹部有用吗？", "predicted": -100, "true_label": "练瑜伽对女人有什么好处？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n尽管存在区域性财务影响，但似乎还有压倒性的社会模式继续这一趋势。 ⇔ ", "response": "尽管存在区域性的财务影响，但社会似乎还有模式继续这项趋势。", "predicted": -100, "true_label": "尽管存在区域性财务影响，但似乎仍有压倒性的社会模式使这一趋势长期存在。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n尽管存在区域性财务影响，但似乎仍有压倒性的社会模式使这一趋势长期存在。 ⇔ ", "response": "尽管存在区域性财务影响，但社会模式使这种趋势持续了很长时间。", "predicted": -100, "true_label": "尽管存在区域性财务影响，但似乎还有压倒性的社会模式继续这一趋势。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA family walking with a soldier. ⇔ ", "response": "A soldier leads a family and two children down a sidewalk.", "predicted": -100, "true_label": "A group of people strolling."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of people strolling. ⇔ ", "response": "A group of people walking down the street.", "predicted": -100, "true_label": "A family walking with a soldier."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天是个晴天吗 ⇔ ", "response": "今天是个晴天吗？", "predicted": -100, "true_label": "今天是个晴天马"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天是个晴天马 ⇔ ", "response": "今天是个晴空万里马", "predicted": -100, "true_label": "今天是个晴天吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党议员和前民主党领袖组成。 ⇔ ", "response": "PLEO通常由��主党全国委员会成员，民主党国会议员，民主党议员和前民主党领袖组成。", "predicted": -100, "true_label": "PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党长和前党派领导人组成。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党长和前党派领导人组成。 ⇔ ", "response": "PLEO代表通常由民主党全国委员会成员，民主党国会议员和民主党州长组成，有时甚至由前党派领导人组成。", "predicted": -100, "true_label": "PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党议员和前民主党领袖组成。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第 46 火箭师于 1941 年 4 月 29 日创立，当时第 188 步兵师已于考纳斯完成组建。 ⇔ ", "response": "第 46 火箭师于 1941 年 4 月 29 日组建，由考纳斯完成第 188 步兵师的组建。", "predicted": 46, "true_label": "第 46 火箭师的历史始于 1941 年 4 月 29 日，当时第 188 步枪师在考纳斯组建完成。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第 46 火箭师的历史始于 1941 年 4 月 29 日，当时第 188 步枪师在考纳斯组建完成。 ⇔ ", "response": "第 46 火箭师的历史始于 1941 年 4 月 29 日，当时第 188 步枪师在考纳斯组建完成，后来该师被重新编为火箭师。", "predicted": 46, "true_label": "第 46 火箭师于 1941 年 4 月 29 日创立，当时第 188 步兵师已于考纳斯完成组建。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of young adults are standing on a street in front of a brick wall with a tall wooden gate. ⇔ ", "response": "The group is at a street corner.", "predicted": -100, "true_label": "A group of friends poses for a picture in front of a brick wall."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of friends poses for a picture in front of a brick wall. ⇔ ", "response": "Friends take a picture outside of a building.", "predicted": -100, "true_label": "A group of young adults are standing on a street in front of a brick wall with a tall wooden gate."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe rhino is grazing on the grass ⇔ ", "response": "A rhino is walking in the field", "predicted": -100, "true_label": "The rhino is being fed on the grass"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe rhino is being fed on the grass ⇔ ", "response": "A rhinoceros is being fed", "predicted": -100, "true_label": "The rhino is grazing on the grass"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a suit driving a horse-drawn buggy down a stone street. ⇔ ", "response": "A man is driving a buggy.", "predicted": -100, "true_label": "The man is driving a buggy."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is driving a buggy. ⇔ ", "response": "A man is driving a one horse car, also known as a buggy.", "predicted": -100, "true_label": "A man in a suit driving a horse-drawn buggy down a stone street."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n向Borsalino致敬的是Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑。 ⇔ ", "response": "向Borsalino致敬的是Flos的Chapeau Lamp（2014年）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑。", "predicted": 2014, "true_label": "由Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑都是对Borsalino的致敬。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n由Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑都是对Borsalino的致敬。 ⇔ ", "response": "由Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）都是对Borsalino的致敬。", "predicted": 2014, "true_label": "向Borsalino致敬的是Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n三个孩子在雪中玩耍。 ⇔ ", "response": "三个孩子都在雪中玩耍。", "predicted": -100, "true_label": "三个孩子在干草里玩耍"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n三个孩子在干草里玩耍 ⇔ ", "response": "三个孩子在一包干草里玩耍。", "predicted": -100, "true_label": "三个孩子在雪中玩耍。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy swims shirtless with a black and red bathing suit while a seagull sits on the sand. ⇔ ", "response": "A boy is swimming.", "predicted": -100, "true_label": "A boy swims."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy swims. ⇔ ", "response": "A boy in a swimsuit body surfing near some rocks.", "predicted": -100, "true_label": "A boy swims shirtless with a black and red bathing suit while a seagull sits on the sand."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你妹妹是一条龙 ⇔ ", "response": "我妹妹是只龙", "predicted": -100, "true_label": "我是一条龙"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我是一条龙 ⇔ ", "response": "我是一条魔法龙", "predicted": -100, "true_label": "你妹妹是一条龙"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe head of the US Democratic Party has lambasted Mankiw, and suggested that he be outsourced to India. ⇔ ", "response": "Mankiw works for the Democratic Party.", "predicted": -100, "true_label": "Jobs were outsourced to India."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nJobs were outsourced to India. ⇔ ", "response": "The switch is part of a trend which Deloitte Research reckons will result in 2m of the 13m jobs in financial services in developed economies moving to India by 2008.", "predicted": 2, "true_label": "The head of the US Democratic Party has lambasted Mankiw, and suggested that he be outsourced to India."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n哥哥姐姐们，拜托了 ⇔ ", "response": "哥哥姐姐们，求你了", "predicted": -100, "true_label": "哥哥姐姐们，拜托了，"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n哥哥姐姐们，拜托了， ⇔ ", "response": "哥哥姐姐们，求求了，我错了，我也想考好，您们相信我，我一定改，别放弃了，我一定行，我一定改，别放弃我，我一定行，求你了，您们相信我，我也相信您们一定能够帮我，别放弃我，我一定改，别放弃我，我一定行，哥哥姐姐们，拜托了。", "predicted": -100, "true_label": "哥哥姐姐们，拜托了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么烧猪蹄好吃呢？ ⇔ ", "response": "烧猪蹄好吃吗？", "predicted": -100, "true_label": "猪蹄炖什么才好吃"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n猪蹄炖什么才好吃 ⇔ ", "response": "猪蹄什么炖最好吃", "predicted": -100, "true_label": "什么烧猪蹄好吃呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\ndrop ⇔ ", "response": "indx", "predicted": -100, "true_label": "probebly"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nprobebly ⇔ ", "response": "probablycorrect", "predicted": -100, "true_label": "drop"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWHATAREYOUGOINGTODO ⇔ ", "response": "Whatareyougoing", "predicted": -100, "true_label": "Whereareyougoing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhereareyougoing ⇔ ", "response": "Whattimeisit", "predicted": -100, "true_label": "WHATAREYOUGOINGTODO"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n梯形的高怎么求 ⇔ ", "response": "梯形的高怎么求", "predicted": -100, "true_label": "第二笔是什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n第二笔是什么 ⇔ ", "response": "第二笔是什么字", "predicted": -100, "true_label": "梯形的高怎么求"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe government thereby perpetuates our continued dependence on fuels that are environmentally and economically unsustainable. ⇔ ", "response": "The government relies on fuels that are environmentally and economically unsustainable.", "predicted": -100, "true_label": "The government took measures to protect the environment."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe government took measures to protect the environment. ⇔ ", "response": "Government subsidies for energy and raw materials held prices at artificially low levels, doing away with incentives to conserve natural resources.", "predicted": -100, "true_label": "The government thereby perpetuates our continued dependence on fuels that are environmentally and economically unsustainable."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天我去上课 ⇔ ", "response": "明天我去上课", "predicted": -100, "true_label": "明天我去上个"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天我去上个 ⇔ ", "response": "明天我去上", "predicted": -100, "true_label": "明天我去上课"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这是一个测试文本 ⇔ ", "response": "这是不是个测试文本", "predicted": -100, "true_label": "这是一亇测试文本"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这是一亇测���文本 ⇔ ", "response": "测 试 这一 晋江文学网", "predicted": -100, "true_label": "这是一个测试文本"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man wearing a hat and boots is digging for something in the snow. ⇔ ", "response": "A man is digging up some sort of item from the snow.", "predicted": -100, "true_label": "A man is trying to get to something by digging."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is trying to get to something by digging. ⇔ ", "response": "A man, wearing a digging mask and protective gear, is digging the soil from a hole to reach something buried there.", "predicted": -100, "true_label": "A man wearing a hat and boots is digging for something in the snow."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA team of men, three in red shirts and two others in baseball jerseys, re-surface a baseball diamond with specialized tools. ⇔ ", "response": "A team prepares a baseball field.", "predicted": -100, "true_label": "A grounds crew maintain a park."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA grounds crew maintain a park. ⇔ ", "response": "A group of grounders are cutting grass and weeds in a park.", "predicted": -100, "true_label": "A team of men, three in red shirts and two others in baseball jerseys, re-surface a baseball diamond with specialized tools."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它由Sakamoto编写，由瑞典歌手Frida Sundemo创作。 ⇔ ", "response": "它由Sakamoto创作，由瑞典创作歌手Frida Sundemo编写。", "predicted": -100, "true_label": "它由Sakamoto创作，由瑞典歌手Frida Sundemo编写。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它由Sakamoto创作，由瑞典歌手Frida Sundemo编写。 ⇔ ", "response": "它由Sakamoto编写，由瑞典创作歌手Frida Sundemo创作。", "predicted": -100, "true_label": "它由Sakamoto编写，由瑞典歌手Frida Sundemo创作。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nllll ⇔ ", "response": "IILII", "predicted": -100, "true_label": "lllll"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nlllll ⇔ ", "response": "ll", "predicted": -100, "true_label": "llll"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTony-award winning dancer and actor Gregory Hines died of cancer Saturday in Los Angeles . ⇔ ", "response": "Tony-award winning dancer and actor Gregory Hines died today of cancer in Los Angeles at the age of 55 .", "predicted": 55, "true_label": "Hines died yesterday in Los Angeles of cancer , publicist Allen Eichorn said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHines died yesterday in Los Angeles of cancer , publicist Allen Eichorn said . ⇔ ", "response": "R&B singer Hines died of cancer at 58 , lawyer says .", "predicted": 58, "true_label": "Tony-award winning dancer and actor Gregory Hines died of cancer Saturday in Los Angeles ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样写好写快钢笔字 ⇔ ", "response": "快钢笔字怎么写", "predicted": -100, "true_label": "怎样写好钢笔字。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样写好钢笔字。 ⇔ ", "response": "钢笔字怎么才能写的好看", "predicted": -100, "true_label": "怎样写好写快钢笔字"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWiFi6 ⇔ ", "response": "WIFI", "predicted": -100, "true_label": "wifi"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nwifi ⇔ ", "response": "WIFI6", "predicted": 6, "true_label": "WiFi6"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天天气tianqi怎么样？ ⇔ ", "response": "明天天气怎么样", "predicted": -100, "true_label": "明天的天气将会很hao。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天的天气将会很hao。 ⇔ ", "response": "明天天气很好。", "predicted": -100, "true_label": "明天天气tianqi怎么样？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is sharpening a knife ⇔ ", "response": "There is no man sharpening a knife", "predicted": -100, "true_label": "A man is sharpening a weapon"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is sharpening a weapon ⇔ ", "response": "A man is shaving", "predicted": -100, "true_label": "A man is sharpening a knife"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两只动物坐在岩石上。 ⇔ ", "response": "猫在盘子里的食物。", "predicted": -100, "true_label": "两只老鹰坐在一根长长的树枝上。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两只老鹰坐在一根长长的树枝上。 ⇔ ", "response": "两只鹰坐在树枝上。", "predicted": -100, "true_label": "���只动物坐在岩石上。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head. ⇔ ", "response": "A man and woman are in a field.", "predicted": -100, "true_label": "A couple are having a photoshoot."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA couple are having a photoshoot. ⇔ ", "response": "A couple posing for a photographer.", "predicted": -100, "true_label": "A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nGenerally the coffee or tea drinker feels worse in just a few short minutes after finishing the drink. ⇔ ", "response": "I drink coffee and tea regularly.", "predicted": -100, "true_label": "Tea protects from illness."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTea protects from illness. ⇔ ", "response": "Although many herbs have been grown for centuries for medicinal use, do not look to herbal teas to remedy illness or disease.", "predicted": -100, "true_label": "Generally the coffee or tea drinker feels worse in just a few short minutes after finishing the drink."}
+{"prompt": "Please generate a paraphrase for the following sentence.\na young boy using a field microscope to identify a field specimen during a field trip. ⇔ ", "response": "A boy is using a microscope.", "predicted": -100, "true_label": "The boy is looking through a microscope."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe boy is looking through a microscope. ⇔ ", "response": "A boy with glasses is looking at a slide with a specimen on it.", "predicted": -100, "true_label": "a young boy using a field microscope to identify a field specimen during a field trip."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOne of the main treatments used to combat the symptoms of mental illness is medication in various forms. ⇔ ", "response": "Drugs are used to combat the symptoms of mental illness.", "predicted": -100, "true_label": "Drugs are used to combat schizophrenia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDrugs are used to combat schizophrenia. ⇔ ", "response": "Schizophrenia is a severe psychiatric illness that affects 1% of people worldwide. Drugs used to combat the condition are less than satisfactory: developed through trial and error, many patients react badly to them and there is a growing body of evidence that they have a range of side-effects, including diabetes.", "predicted": 1, "true_label": "One of the main treatments used to combat the symptoms of mental illness is medication in various forms."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n北京现代怎么样？ ⇔ ", "response": "北京现代车怎么样？", "predicted": -100, "true_label": "北京现代怎么样"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n北京现代怎么样 ⇔ ", "response": "北京现代车怎么样", "predicted": -100, "true_label": "北京现代怎么样？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机充电后自动关机开机后就一直打不开 ⇔ ", "response": "电脑充电自动关机", "predicted": -100, "true_label": "手机没电关机后充电一直震动无法开机"}

eval_gen/gpu_results/results_gpu_5.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机没电关机后充电一直震动无法开机 ⇔ ", "response": "充电器连接好手机却无法开机", "predicted": -100, "true_label": "手机充电后自动关机开机后就一直打不开"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMansfield attempted to list every vessel that ever sailed the Lakes before 1899, as well as every shipwreck. ⇔ ", "response": "A ship went down in the Lakes.", "predicted": -100, "true_label": "Shipwreck salvaging was attempted."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nShipwreck salvaging was attempted. ⇔ ", "response": "Hundreds of divers and treasure hunters, including the Duke of Argyll, have risked their lives in the dangerous waters of the Isle of Mull trying to discover the reputed 30,000,000 pounds in Gold carried by this vessel--the target of the most enduring treasure hunt in British history.", "predicted": 30, "true_label": "Mansfield attempted to list every vessel that ever sailed the Lakes before 1899, as well as every shipwreck."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA shirtless man is painting over a graffiti filled wall with a paint roller. ⇔ ", "response": "A man is a painting the wall", "predicted": -100, "true_label": "A man is painting a wall."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is painting a wall. ⇔ ", "response": "A man paints a room.", "predicted": -100, "true_label": "A shirtless man is painting over a graffiti filled wall with a paint roller."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n叙利亚要求保证，反政府武装称和平计划注定要失败。 ⇔ ", "response": "叙利亚总统表示，反对派已誓言参与和平进程。", "predicted": -100, "true_label": "叙利亚要求提供撤军的书面保证"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n叙利亚要求提供撤军的书面保证 ⇔ ", "response": "叙利亚拒绝在撤军中引入伊朗", "predicted": -100, "true_label": "叙利亚要求保证，反政府武装称和平计划注定要失败。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is riding a horse in a sandy land ⇔ ", "response": "A person is riding a horse", "predicted": -100, "true_label": "A guy is riding a horse"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA guy is riding a horse ⇔ ", "response": "A man is riding a horse", "predicted": -100, "true_label": "A man is riding a horse in a sandy land"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n离这儿不远有一座寺庙，表弟说，里面住着一位修士，据说，此人当过兵，口碑不错，而且还非常谦和善良。 ⇔ ", "response": "这儿不远的寺庙里住着一位修士，这位修士当过兵，口碑不错，而且还非常谦和善良。", "predicted": -100, "true_label": "那位表亲说:离这儿不远，住着个隐居的修士，据说当过兵，大家认为他是个很有见识的好基督徒，对人也很诚实。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n那位表亲说:离这儿不远，住着个隐居的修士，据说当过兵，大家认为他是个很有见识的好基督徒，对人也很诚实。 ⇔ ", "response": "隐居的修士，大家认为他是个很有见识的好基督徒，也对人很诚实。", "predicted": -100, "true_label": "离这儿不远有一座寺庙，表弟说，里面住着一位修士，据说，此人当过兵，口碑不错，而且还非常谦和善良。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在线宠物医生咨询 ⇔ ", "response": "宠物医生咨询时间", "predicted": -100, "true_label": "在线宠物医生咨询急！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在线宠物医生咨询急！ ⇔ ", "response": "在线电脑咨询急！", "predicted": -100, "true_label": "在线宠物医生咨询"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好啊 ⇔ ", "response": "今天天气真好a", "predicted": -100, "true_label": "今天天啊好真"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天啊好真 ⇔ ", "response": "今天天气真好啊", "predicted": -100, "true_label": "今天天气真好啊"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样关闭防止游戏提示 ⇔ ", "response": "怎么让游戏不提示已解锁", "predicted": -100, "true_label": "总提示软件有危险怎么关闭"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n总提示软件有危险怎么关闭 ⇔ ", "response": "软件提示有危险如何取消提示", "predicted": -100, "true_label": "怎样关闭防止游戏提示"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n穿绿色运动衫的男孩在足球队担任守门员。 ⇔ ", "response": "穿着绿色运动衫的男孩在足球队担任门将。", "predicted": -100, "true_label": "穿蓝衣服的球队正在和穿白色衣服的球队比赛。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n穿蓝衣服的球队正在和穿白色衣服的球队比赛。 ⇔ ", "response": "蓝衣队和白衣队正在比赛。", "predicted": -100, "true_label": "穿绿色运动衫的男孩在足球队担任守门员。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAnd his justification for the bombing was that while it caused short-term material damage , it was for the long-term moral good of Bali . ⇔ ", "response": "His justification for the bombing was that it caused short-term material damage but was necessary for the long-term moral good of Bali .", "predicted": -100, "true_label": "And he justified bombing Bali by saying that while it had caused material devastation , it was for the island 's long-term moral good ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAnd he justified bombing Bali by saying that while it had caused material devastation , it was for the island 's long-term moral good . ⇔ ", "response": "He justified the bombing campaign by saying it caused material devastation but was necessary for Balinese long-term moral good .", "predicted": -100, "true_label": "And his justification for the bombing was that while it caused short-term material damage , it was for the long-term moral good of Bali ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有几个可能的原因。 ⇔ ", "response": "原因可能有好几个。", "predicted": -100, "true_label": "画皱纹有几个原因。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n画皱纹有几个原因。 ⇔ ", "response": "画皱纹是因为它是生命的证据。", "predicted": -100, "true_label": "有几个可能的原因。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两个女人站在旅游车前。 ⇔ ", "response": "两个女人，一个戴着太阳镜，一个戴着帽子，站在旅游车前。", "predicted": -100, "true_label": "两个人在飞机前说话。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两个人在飞机前说话。 ⇔ ", "response": "两个人在飞机上说话。", "predicted": -100, "true_label": "两个女人站在旅游车前。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机录屏黑屏是什么原因 ⇔ ", "response": "手机录屏常黑屏是什么原因", "predicted": -100, "true_label": "手机黑屏关机是什么原因"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机黑屏关机是什么原因 ⇔ ", "response": "手机蓝屏关机是什么原因", "predicted": -100, "true_label": "手机录屏黑屏是什么原因"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n人们试图将修道院的纳扎尔人解释为对禁欲训练实践上瘾的圣经命令的先驱。 ⇔ ", "response": "人们试图将纳扎尔人修道院解释为圣经中追求简单生活方式的前沿的瘾君子训练实践。", "predicted": -100, "true_label": "已经尝试解释修道士纳扎尔人作为圣经命令的先驱，他们沉迷于禁欲训练的实践。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n已经尝试解释修道士纳扎尔人作为圣经命令的先驱，他们沉迷于禁欲训练的实践。 ⇔ ", "response": "已经尝试解释修道士纳扎尔人作为圣经命令的先驱，他们沉迷于禁欲训练的实践。", "predicted": -100, "true_label": "人们试图将修道院的纳扎尔人解释为对禁欲训练实践上瘾的圣经命令的先驱。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAksuat Dendropark（俄语：Akshut Dendroparks）是一座受保护的纪念碑（Ulyanovsk Oblast的自然区域）。 ⇔ ", "response": "Aksuat Dendropark（俄语：Акшуат Дендропарк）是一座受保护的纪念碑（Ulyanovsk Oblast的自然区域）。", "predicted": -100, "true_label": "Aksuat Dendropark（俄语：Akshout Dendroparks）是一个天然纪念物（乌里扬诺夫斯克州保护区）"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAksuat Dendropark（俄语：Akshout Dendroparks）是一个天然纪念物（乌里扬诺夫斯克州保护区） ⇔ ", "response": "Aksuat Dendropark是一个天然纪念物（乌里扬诺夫斯克州保护区）", "predicted": -100, "true_label": "Aksuat Dendropark（俄语：Akshut Dendroparks）是一座受保护的纪念碑（Ulyanovsk Oblast的自然区域）。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRevenues for \" The Hulk \" came in well below those of last month 's Marvel Comics adaptation , \" X2 : X-Men United , \" which grossed $ 85.6 million in its opening weekend . ⇔ ", "response": "\" The Hulk \" came in at $ 50 million , well below \" X2 : X-Men United \"s $ 85.6 million opening ; that film was a Marvel Comics adaptation .", "predicted": 50, "true_label": "The Hulk trailed last month 's Marvel Comics adaptation , X2 : X-Men United , which grossed $ 85.6-million in its opening weekend ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe Hulk trailed last month 's Marvel Comics adaptation , X2 : X-Men United , which grossed $ 85.6-million in its opening weekend . ⇔ ", "response": "The Hulk on Saturday fell back to second place in its second week , trailing \" X2 : X-Men United ,\" which again \" Marvel Comics \" , opened last weekend , grossed $ 85.6 million in its opening weekend .", "predicted": 2, "true_label": "Revenues for \" The Hulk \" came in well below those of last month 's Marvel Comics adaptation , \" X2 : X-Men United , \" which grossed $ 85.6 million in its opening weekend ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" These foods have an almost identical effect on lowering cholesterol as the original cholesterol-lowering drugs . \" ⇔ ", "response": "They will have an almost identical effect on lowering cholesterol , pma ii chairman dr . richard posen said at a news conference .", "predicted": -100, "true_label": "We have now proven that these foods have an almost identical effect on lowering cholesterol as the original cholesterol-reducing drugs ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWe have now proven that these foods have an almost identical effect on lowering cholesterol as the original cholesterol-reducing drugs . ⇔ ", "response": "This shows that the foods have an almost identical effect on lowering cholesterol as the original drugs .", "predicted": -100, "true_label": "\" These foods have an almost identical effect on lowering cholesterol as the original cholesterol-lowering drugs . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man on a motorcycle is riding on one wheel ⇔ ", "response": "A man is riding a motorcycle", "predicted": -100, "true_label": "A person is performing acrobatics on a motorcycle"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is performing acrobatics on a motorcycle ⇔ ", "response": "A person is riding a motorcycle while doing an acrobatic move", "predicted": -100, "true_label": "A man on a motorcycle is riding on one wheel"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n电影中的岛屿位于埃尔斯米尔岛以北，而不是帕特里克王子岛。 ⇔ ", "response": "电影中的岛屿位于埃尔斯米尔岛以北，而不是帕特里克王子岛。", "predicted": -100, "true_label": "该岛屿位于埃尔斯米尔岛以北的电影中，而不是帕特里克王子岛。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该岛屿位于埃尔斯米尔岛以北的电影中，而不是帕特里克王子岛。 ⇔ ", "response": "电影中，它位于埃尔斯米尔岛以北的一块土地实际上是一个帕特里克王子岛。", "predicted": -100, "true_label": "电影中的岛屿位于埃尔斯米尔岛以北，而不是帕特里克王子岛。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOppoa7 ⇔ ", "response": "opoopa7", "predicted": 7, "true_label": "opoopa73"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nopoopa73 ⇔ ", "response": "Oppoa8", "predicted": 8, "true_label": "Oppoa7"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of men is playing soccer on the beach ⇔ ", "response": "People are playing soccer in the sea", "predicted": -100, "true_label": "A group of men is playing with a ball on the beach"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of men is playing with a ball on the beach ⇔ ", "response": "A group of men is playing soccer on the beach", "predicted": -100, "true_label": "A group of men is playing soccer on the beach"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打麻将怎么赢钱 ⇔ ", "response": "打麻将赢钱方法", "predicted": -100, "true_label": "打麻将怎么才能赢钱？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打麻将怎么才能赢钱？ ⇔ ", "response": "怎么玩麻将才能赢", "predicted": -100, "true_label": "打麻将怎么赢钱"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n关于夏天的谚语 ⇔ ", "response": "有关夏天的诗句", "predicted": -100, "true_label": "关于读书的名人名言"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n关于读书的名人名言 ⇔ ", "response": "说给老人听的诗", "predicted": -100, "true_label": "关于夏天的谚语"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天氣真好，我們去公園散步吧 ⇔ ", "response": "今天天气真好，我们去公园散步吧", "predicted": -100, "true_label": "天氣真號，我們去公園散步吧"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天氣真號，我們去公園散步吧 ⇔ ", "response": "今天天气很好，我们去公园散步吧", "predicted": -100, "true_label": "今天天氣真好，我們去公園散步吧"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWHATSWRONG ⇔ ", "response": "Whatwhatwhat", "predicted": -100, "true_label": "Whatswrongwithyou"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhatswrongwithyou ⇔ ", "response": "Donttellmewhattodo", "predicted": -100, "true_label": "WHATSWRONG"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一年好景君须记，下一句是什么 ⇔ ", "response": "一年好景君须记，莫使金樽空对月", "predicted": -100, "true_label": "一年好景君须记的下一句是什么？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一年好景君须记的下一句是什么？ ⇔ ", "response": "一年好景君须记", "predicted": -100, "true_label": "一年好景君须记，下一句是什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman slicing an onion ⇔ ", "response": "A woman is slicing an onion", "predicted": -100, "true_label": "A woman is slicing an onion"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is slicing an onion ⇔ ", "response": "There is no woman slicing an onion", "predicted": -100, "true_label": "There is no woman slicing an onion"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你好小哦 ⇔ ", "response": "你好小哟", "predicted": -100, "true_label": "你的小仓鼠呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你的小仓鼠呢 ⇔ ", "response": "仓鼠住在哪个房间", "predicted": -100, "true_label": "你好小哦"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman looks at a plate filled with steam. ⇔ ", "response": "A woman in a dark shirt is looking at a plate that is filled with steam.", "predicted": -100, "true_label": "The woman is in a kitchen."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is in a kitchen. ⇔ ", "response": "A woman in a striped shirt and a slacks skirt is in the kitchen.", "predicted": -100, "true_label": "A woman looks at a plate filled with steam."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n外祖母满心感激地说。 ⇔ ", "response": "外祖母心里感激不尽，脸上却露出了痛苦难忍的表情。", "predicted": -100, "true_label": "外祖母感激地说。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n外祖母感激地说。 ⇔ ", "response": "外祖母感激地一笑。", "predicted": -100, "true_label": "外祖母满心感激地说。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe nightmare of human disaster in Rwanda in 1994, after UN Peacekeeping Forces withdrew, shocked us into questioning the whole concept of peacekeeping. ⇔ ", "response": "Peacekeeping forces were established.", "predicted": -100, "true_label": "UN peacekeeping forces withdrew from Bosnia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nUN peacekeeping forces withdrew from Bosnia. ⇔ ", "response": "The decision to withdraw comes as Bosnia holds its first general elections since the 1995-1996 war.", "predicted": 1995, "true_label": "The nightmare of human disaster in Rwanda in 1994, after UN Peacekeeping Forces withdrew, shocked us into questioning the whole concept of peacekeeping."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她们两个有时单人独唱，有时二人合唱，我们很可以算得开了一个小小的音乐会。 ⇔ ", "response": "她们两个有时单人独唱，有时二人合唱，我们很可以算开了一个小小的音乐会。", "predicted": -100, "true_label": "她俩一起唱，还进行二声部合唱，这一来我们就举行了一个很圆满的小型音乐会。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她俩一起唱，还进行二声部合唱，这一来我们就举行了一个很圆满的小型音乐会。 ⇔ ", "response": "她们俩一起唱，还进行二声部合唱，这就使我们得到一次非常圆满的小型音乐会。", "predicted": -100, "true_label": "她们两个有时单人独唱，有时二人合唱，我们很可以算得开了一个小小的音乐会。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nFive people standing on a stage, getting ready to perform, who are unrecognizable due to the lack of lighting. ⇔ ", "response": "Five people are about to perform, but cannot be seen due to poor lighting.", "predicted": -100, "true_label": "In a dim room, several actors who can barely see each other, are getting ready to act."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIn a dim room, several actors who can barely see each other, are getting ready to act. ⇔ ", "response": "The actors are getting ready.", "predicted": -100, "true_label": "Five people standing on a stage, getting ready to perform, who are unrecognizable due to the lack of lighting."}
+{"prompt": "Please generate a paraphrase for the following sentence.\na soccer player with black and red striped uniform gets ready to kick the soccer ball ⇔ ", "response": "A soccer player is getting ready to kick the ball", "predicted": -100, "true_label": "a soccer player gets ready to kick the ball after a penalty."}
+{"prompt": "Please generate a paraphrase for the following sentence.\na soccer player gets ready to kick the ball after a penalty. ⇔ ", "response": "A soccer player is preparing to score a goal.", "predicted": -100, "true_label": "a soccer player with black and red striped uniform gets ready to kick the soccer ball"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它以Robert Suhosky的小说为基础，并由James Hardiman改编成剧本。 ⇔ ", "response": "它以Robert Suhosky的小说为基础，并由James Hardiman改编成剧本。", "predicted": -100, "true_label": "它是基于罗伯特·苏霍斯基的一部小说改编成詹姆斯·哈迪曼的剧本。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n它是基于罗伯特·苏霍斯基的一部小说改编成詹姆斯·哈迪曼的剧本。 ⇔ ", "response": "它基于罗伯特·苏霍斯基的小说编写了詹姆斯·哈迪曼的剧本。", "predicted": -100, "true_label": "它以Robert Suhosky的小说为基础，并由James Hardiman改编成剧本。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的手机在阳光下屏一点看不见 ⇔ ", "response": "我的手机在阳光下屏幕看不见了怎么办", "predicted": -100, "true_label": "我的手机屏幕碎了打开后也点不开"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的手机屏幕碎了打开后也点不开 ⇔ ", "response": "我的手机屏幕碎了打开后也点不开了吗", "predicted": -100, "true_label": "我的手机在阳光下屏一点看不见"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n最后，在1601年，出版了委托散文故事（“Bataviae Hollandiaeque Annales”）。 ⇔ ", "response": "最后，在1601年，出版了委托的叙述性散文故事（“荷兰弗雷德里卡二世在巴达维亚的年表”）。", "predicted": 1601, "true_label": "最后，出版的散文历史（“Bataviae Hollandiaeque Annales”）于1601年投入使用。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n最后，出版的散文历史（“Bataviae Hollandiaeque Annales”）于1601年投入使用。 ⇔ ", "response": "最后，出版于1601年的散文历史（“Bataviae Hollandiaeque Annales”）用法语完成。", "predicted": 1601, "true_label": "最后，在1601年，出版了委托散文故事（“Bataviae Hollandiaeque Annales”）。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怀孕能吃毛笋吗 ⇔ ", "response": "怀孕能吃毛笋吗？", "predicted": -100, "true_label": "泥鳅吃什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n泥鳅吃什么 ⇔ ", "response": "泥鳅喜欢什么", "predicted": -100, "true_label": "怀孕能吃毛笋吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nViolin soloists take the stage during the orchestra's opening show at the theater. ⇔ ", "response": "a violin soloist performs on stage", "predicted": -100, "true_label": "People are playing violin on stage."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople are playing violin on stage. ⇔ ", "response": "A person is playing the violin.", "predicted": -100, "true_label": "Violin soloists take the stage during the orchestra's opening show at the theater."}
+{"prompt": "Please generate a paraphrase for the following sentence.\npeople standing at a beach with Cameras. ⇔ ", "response": "A group of people take photos at a beach.", "predicted": -100, "true_label": "A group of people standing at a beach filled with cameras."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of people standing at a beach filled with cameras. ⇔ ", "response": "People are standing at a beach.", "predicted": -100, "true_label": "people standing at a beach with Cameras."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIraq 's nuclear program had been dismantled and there was no convincing evidence it was being revived , the report said . ⇔ ", "response": "The report said Iraq had kept its program dismantled and there was no convincing evidence it was being revived .", "predicted": -100, "true_label": "Iraq 's nuclear program had been dismantled , and there \" was no convincing evidence of its reconstitution . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIraq 's nuclear program had been dismantled , and there \" was no convincing evidence of its reconstitution . \" ⇔ ", "response": "Saddam had dismantled Iraq 's nuclear program , and there was \" no convincing evidence of its reconstitution . \"", "predicted": -100, "true_label": "Iraq 's nuclear program had been dismantled and there was no convincing evidence it was being revived , the report said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n四只黑狗跑过草地。 ⇔ ", "response": "四只黑狗在草地上奔跑。", "predicted": -100, "true_label": "一只狗跑过草地。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一只狗跑过草地。 ⇔ ", "response": "一只棕色的狗跑过草地，身上带着一根绳子。", "predicted": -100, "true_label": "四只黑狗跑过草地。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA cat is sliding upside down under a couch ⇔ ", "response": "A big cat is sliding under a couch", "predicted": -100, "true_label": "An animal is crawling under a piece of furniture"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn animal is crawling under a piece of furniture ⇔ ", "response": "A snake is crawling under a piece of furniture", "predicted": -100, "true_label": "A cat is sliding upside down under a couch"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n越南探测失踪飞机信号 ⇔ ", "response": "搜索团队在澳大利亚失踪飞机上找到残骸", "predicted": -100, "true_label": "中国扩大失踪飞机搜寻范围"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国扩大失踪飞机搜寻范围 ⇔ ", "response": "中国扩大失踪飞机搜寻面积", "predicted": -100, "true_label": "越南探测失踪飞机信号"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在数学天文学领域，他的声誉归功于天文球体的提出，以及他对理解行星运动的早期贡献。 ⇔ ", "response": "在数学天文学领域，他的声誉归功于他对天文球体的提出以及他早期对理解行星运动的贡献。", "predicted": -100, "true_label": "他在数学天文学方面享有盛誉是因为他引入了天文地球仪，并对理解行星运动作出了早期贡献。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他在数学天文学方面享有盛誉是因为他引入了天文地球仪，并对理解行星运动作出了早期贡献。 ⇔ ", "response": "他因引入天文地球仪和早期理解行星运动而闻名于世，地球仪是一种数学天文学。", "predicted": -100, "true_label": "在数学天文学领域，他的声誉归功于天文球体的提出，以及他对理解行星运动的早期贡献。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这个美女叫什么名字 ⇔ ", "response": "请问这个叫什么", "predicted": -100, "true_label": "请问这个美女叫什么名字！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请问这个美女叫什么名字！ ⇔ ", "response": "请问这个叫什么名字", "predicted": -100, "true_label": "请问这个美女叫什么名字"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of children divided into a maroon team and a blue team playing a soccer game. ⇔ ", "response": "A group of children are playing soccer.", "predicted": -100, "true_label": "Kids are playing soccer."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nKids are playing soccer. ⇔ ", "response": "Many CHANEL bags would make appearances on the field, but it's the simple white soccer ball that dominates the space, a white, leather ball that is both the game's objective and the focal point of the scene.", "predicted": -100, "true_label": "A group of children divided into a maroon team and a blue team playing a soccer game."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天去公园玩了。 ⇔ ", "response": "我昨天去公园玩了。", "predicted": -100, "true_label": "我公元去玩昨天了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我公元去玩昨天了。 ⇔ ", "response": "昨天我去玩了。", "predicted": -100, "true_label": "我昨天去公园玩了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜欢打篮球。 ⇔ ", "response": "我喜欢打足球。", "predicted": -100, "true_label": "我喜欢打蓝球。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜欢打蓝球。 ⇔ ", "response": "我喜欢打篮球。", "predicted": -100, "true_label": "我喜欢打篮球。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n20世纪90年代，在亚历山德里亚，比耶拉，伊夫雷亚，蒙多维和韦尔切利开设了新的教学场所。 ⇔ ", "response": "20世纪90年代，在亚历山德里亚，比耶拉，伊夫雷亚，蒙多维和韦尔切利开设了新的教学设施。", "predicted": 20, "true_label": "在20世纪90年代，在Ivrea，Mondovì，Biella，Alessandria和Vercelli开设了新的教学园。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在20世纪90年代，在Ivrea，Mondovì，Biella，Alessandria和Vercelli开设了新的教学园。 ⇔ ", "response": "在20世纪90年代，在Mondovì，Ivrea，Biella，Alessandria和Vercelli开设了新的教学园区。", "predicted": 20, "true_label": "20世纪90年代，在亚历山德里亚，比耶拉，伊夫雷亚，蒙多维和韦尔切利开设了新的教学场所。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nJack Ferry , company spokesman , said a search is ongoing but would not comment on the status . ⇔ ", "response": "Jack Ferry , a company spokesman , said Tuesday morning that a search is ongoing but would not comment on the status of the passengers .", "predicted": -100, "true_label": "Company spokesman Jack Ferry said a search is ongoing but declined comment on its status ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCompany spokesman Jack Ferry said a search is ongoing but declined comment on its status . ⇔ ", "response": "A search is under way for the aircraft , and company spokesman Jack Ferry said he could not yet comment on its status .", "predicted": -100, "true_label": "Jack Ferry , company spokesman , said a search is ongoing but would not comment on the status ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is not riding a mechanical bull ⇔ ", "response": "The man is riding a mechanical bull", "predicted": -100, "true_label": "The man is riding a mechanical bull"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is riding a mechanical bull ⇔ ", "response": "The man is riding a pony", "predicted": -100, "true_label": "The man is not riding a mechanical bull"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe Government's new policies give more weight to reducing the adverse impact of trunk roads on the environment, including reducing greenhouse gas emissions and pollutants from trunk road traffic. ⇔ ", "response": "The Government policies are focused on reducing emissions.", "predicted": -100, "true_label": "The government took measures to reduce pollution."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe government took measures to reduce pollution. ⇔ ", "response": "This industrial spills problem has persisted for years and it is about time that government took decisive action, says Paul Muldoon, Executive Director of the Canadian Environmental Law Association.", "predicted": -100, "true_label": "The Government's new policies give more weight to reducing the adverse impact of trunk roads on the environment, including reducing greenhouse gas emissions and pollutants from trunk road traffic."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去商店买了许多食物。 ⇔ ", "response": "我今天去商店买了很多食物。", "predicted": -100, "true_label": "我今天去商店买了许多试物。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去商店买了许多试物。 ⇔ ", "response": "我今天去商店买了许多东西。", "predicted": -100, "true_label": "我今天去商店买了许多食物。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我想去公园散步。 ⇔ ", "response": "今天天气真好，我想去公园散步。", "predicted": -100, "true_label": "今天天气真好，我想去公元散步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我想去公元散步。 ⇔ ", "response": "今天天气真好，我想去公园散步。", "predicted": -100, "true_label": "今天天气真好，我想去公园散步。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAustralia , Chile , Colombia , El Salvador , Honduras , Mexico , New Zealand , Peru and Uruguay will also support the challenge . ⇔ ", "response": "The nine countries supporting the challenge are Australia , Chile , Colombia , El Salvador , Honduras , Mexico , New Zealand , Peru and Uruguay .", "predicted": -100, "true_label": "Nine other countries , including Australia , Chile , Colombia , El Salvador and Mexico , are supporting the case ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNine other countries , including Australia , Chile , Colombia , El Salvador and Mexico , are supporting the case . ⇔ ", "response": "Australia , Chile , Colombia , El Salvador , Mexico , New Zealand , Nicaragua , Peru and Uruguay are supporting the case .", "predicted": -100, "true_label": "Australia , Chile , Colombia , El Salvador , Honduras , Mexico , New Zealand , Peru and Uruguay will also support the challenge ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is preparing an egg in a skillet ⇔ ", "response": "A woman is cracking an egg into a skillet", "predicted": -100, "true_label": "There is no woman preparing an egg in a skillet"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no woman preparing an egg in a skillet ⇔ ", "response": "A woman is preparing an egg in a skillet", "predicted": -100, "true_label": "A woman is preparing an egg in a skillet"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA small guinea pig is gnawing and eating a piece of carrot on the floor ⇔ ", "response": "A guinea pig is eating a carrot", "predicted": -100, "true_label": "There is no small guinea pig gnawing and eating a piece of carrot on the floor"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no small guinea pig gnawing and eating a piece of carrot on the floor ⇔ ", "response": "A small guinea pig is gnawing and eating a piece of carrot on the floor", "predicted": -100, "true_label": "A small guinea pig is gnawing and eating a piece of carrot on the floor"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n频频什么意思 ⇔ ", "response": "频频提示", "predicted": -100, "true_label": "频频是什么意思？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n频频是什么意思？ ⇔ ", "response": "频频意思是经常", "predicted": -100, "true_label": "频频什么意思"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is not eating a banana by a tree ⇔ ", "response": "A man is eating a banana by a tree", "predicted": -100, "true_label": "The man is eating a banana by a tree"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is eating a banana by a tree ⇔ ", "response": "A man is eating a banana", "predicted": -100, "true_label": "The man is not eating a banana by a tree"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你爸爸做什么的 ⇔ ", "response": "你现在几岁了呢", "predicted": -100, "true_label": "打我爸爸"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打我爸爸 ⇔ ", "response": "倒我爸爸", "predicted": -100, "true_label": "你爸爸做什么的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of people in a room, most with computers, some raising their hands. ⇔ ", "response": "A group of people in a room, most with computers, some raising their hands.", "predicted": -100, "true_label": "Some people have questions."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome people have questions. ⇔ ", "response": "A man in a black suit and black tie raises his hand as the person speaking address to him.", "predicted": -100, "true_label": "A group of people in a room, most with computers, some raising their hands."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn eggplant is being sliced by a woman ⇔ ", "response": "There is no eggplant being sliced by a woman", "predicted": -100, "true_label": "A woman is cutting a vegetable"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is cutting a vegetable ⇔ ", "response": "A woman is cutting a tomato", "predicted": -100, "true_label": "An eggplant is being sliced by a woman"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMrs. Clinton said she was incredulous that he would endanger their marriage and family . ⇔ ", "response": "\" I find it hard to believe that he would do anything that would endanger our marriage , our family , \" Mrs. Clinton said .", "predicted": -100, "true_label": "She hadn 't believed he would jeopardize their marriage and family ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nShe hadn 't believed he would jeopardize their marriage and family . ⇔ ", "response": "She had not believed he would risk their marriage and family .", "predicted": -100, "true_label": "Mrs. Clinton said she was incredulous that he would endanger their marriage and family ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is stirring vegetables in a pot ⇔ ", "response": "A person is not stirring vegetables in a pot", "predicted": -100, "true_label": "A person is stirring vegetables in a bowl"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is stirring vegetables in a bowl ⇔ ", "response": "A person is adding ingredients to a bowl", "predicted": -100, "true_label": "A person is stirring vegetables in a pot"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打开同程旅行我要订下周一西安的酒店 ⇔ ", "response": "打开携程我要订西安酒店", "predicted": -100, "true_label": "打开黄瓜视频"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打���黄瓜视频 ⇔ ", "response": "打开小视频播放列表", "predicted": -100, "true_label": "打开同程旅行我要订下周一西安的酒店"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nJapan and Russia reportedly have agreed to study the possibility of constructing an oil pipeline connecting Angarsk in eastern Siberia to Russia's Pacific Coast port of Nakhodka. ⇔ ", "response": "A pipeline was constructed in Siberia.", "predicted": -100, "true_label": "An oil pipeline in Siberia was ruptured."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn oil pipeline in Siberia was ruptured. ⇔ ", "response": "On September 10, 700 tonnes of oil spilled into the Vatyaun and Pim rivers in Russia's Western Siberia due to a local pipeline rupture.", "predicted": 10, "true_label": "Japan and Russia reportedly have agreed to study the possibility of constructing an oil pipeline connecting Angarsk in eastern Siberia to Russia's Pacific Coast port of Nakhodka."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n“粉碎者”于 1987 年 6 月 13 日在日本发行，并由东宝株式会社出版。 ⇔ ", "response": "粉碎者”于 1987 年 6 月 13 日在日本发行，并由东宝株式会社出版。", "predicted": 1987, "true_label": "“粉碎者”于 1987 年 6 月 13 日在日本发布，由东宝发行。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n“粉碎者”于 1987 年 6 月 13 日在日本发布，由东宝发行。 ⇔ ", "response": "“粉碎者”于 1987 年 6 月 13 日在日本发行。", "predicted": 1987, "true_label": "“粉碎者”于 1987 年 6 月 13 日在日本发行，并由东宝株式会社出版。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAt the recent 13th Annual Medicine Meets Virtual Reality Conference in Long Beach, Calif., researchers discussed the development of new technologies using virtual reality to treat soldiers returning from Iraq with posttraumatic stress disorder and to provide those being deployed with stress inoculation training. ⇔ ", "response": "Soldiers were exposed to virtual reality.", "predicted": -100, "true_label": "Soldiers are trained using virtual reality."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSoldiers are trained using virtual reality. ⇔ ", "response": "Virtual reality has many applications. Today the military uses this technology in simulations. Soldiers are trained to fly airplanes, drive tanks and fight wars without having to physically be in danger.", "predicted": -100, "true_label": "At the recent 13th Annual Medicine Meets Virtual Reality Conference in Long Beach, Calif., researchers discussed the development of new technologies using virtual reality to treat soldiers returning from Iraq with posttraumatic stress disorder and to provide those being deployed with stress inoculation training."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n求大神解答啊 ⇔ ", "response": "请问什么答案", "predicted": -100, "true_label": "求大神解答么么嗒"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n求大神解答么么嗒 ⇔ ", "response": "求解", "predicted": -100, "true_label": "求大神解答啊"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这张照片是哪个动漫？ ⇔ ", "response": "这张照片是动漫吗？", "predicted": -100, "true_label": "这张照片是哪个动漫的"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这张照片是哪个动漫的 ⇔ ", "response": "这张照片是动漫人物还是的", "predicted": -100, "true_label": "这张照片是哪个动漫？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎样在微信上卖东西？ ⇔ ", "response": "如何能在微信上发布文章", "predicted": -100, "true_label": "怎么在微信上卖东西"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么在微信上卖东西 ⇔ ", "response": "怎么把东西放在微信上卖", "predicted": -100, "true_label": "怎样在微信上卖东西？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is erasing the other man 's work from the board ⇔ ", "response": "One man is erasing the work of another man on a construction site", "predicted": -100, "true_label": "The man is erasing the other man 's work from the panel"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is erasing the other man 's work from the panel ⇔ ", "response": "One man is removing another man 's painting from a wall", "predicted": -100, "true_label": "The man is erasing the other man 's work from the board"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n1990年至1992年，Walt Harris在White的Division I-A（现为Division I FBS）Pacific担任相同职位。 ⇔ ", "response": "1990年至1992年，Walt Harris在White的Division I-A（现为Division I FBS）Pacific担任相同职位。", "predicted": 1990, "true_label": "从1990年到1992年，沃尔特哈里斯在怀特的第一分部（现为第一分部FBS）太平洋分别担任过同样的职务。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n从1990年到1992年，沃尔特哈里斯在怀特的第一分部（现为第一分部FBS）太平洋分别担任过同样的职务。 ⇔ ", "response": "从1990年到1992年，沃尔特哈里斯在怀特的第一分部太平洋分别担任助理教练（现为FBS第一分部）。", "predicted": 1990, "true_label": "1990年至1992年，Walt Harris在White的Division I-A（现为Division I FBS）Pacific担任相同职位。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天气真好啊，阳光明媚 ⇔ ", "response": "今天天气真好，阳光明媚啊", "predicted": -100, "true_label": "天气真好啊，阳关明媚"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n天气真好啊，阳关明媚 ⇔ ", "response": "天气真好啊，阳光明媚", "predicted": -100, "true_label": "天气真好啊，阳光明媚"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nVillagers pulling in the fish boat. ⇔ ", "response": "The fishers are pulling in their catch.", "predicted": -100, "true_label": "There are people pulling a boat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are people pulling a boat. ⇔ ", "response": "A group of friends are pulling a person in a canoe.", "predicted": -100, "true_label": "Villagers pulling in the fish boat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people, one wearing an orange jersey and a white belt, and another wearing a blue jersey and a red belt are playing flag football. ⇔ ", "response": "Two people are playing flag football.", "predicted": -100, "true_label": "Two people on oposite teams play flag football."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people on oposite teams play flag football. ⇔ ", "response": "Two football players, one for each team, are catching flags.", "predicted": -100, "true_label": "Two people, one wearing an orange jersey and a white belt, and another wearing a blue jersey and a red belt are playing flag football."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n船在波涛汹涌的水中航行。 ⇔ ", "response": "蓝色圆圈图案由多个小圆圈组成。", "predicted": -100, "true_label": "一个人在波涛汹涌的海面上划皮艇。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个人在波涛汹涌的海面上划皮艇。 ⇔ ", "response": "一个人在水上滑板。", "predicted": -100, "true_label": "船在波涛汹涌的水中航行。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman in a striped shirt holds an infant. ⇔ ", "response": "A woman holding an infant.", "predicted": -100, "true_label": "A woman is dressed."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is dressed. ⇔ ", "response": "A woman is putting clothes on.", "predicted": -100, "true_label": "A woman in a striped shirt holds an infant."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个日历还有一只蜜蜂是什么成语 ⇔ ", "response": "一个日历还有一天是什么成语", "predicted": -100, "true_label": "日历表2015日历一只蜜蜂是什么成语"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n日历表2015日历一只蜜蜂是什么成语 ⇔ ", "response": "2015日历一张什么成语", "predicted": 2015, "true_label": "一个日历还有一只蜜蜂是什么成语"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOz ⇔ ", "response": "oppoA", "predicted": -100, "true_label": "oppogztr"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppogztr ⇔ ", "response": "OPPOtr", "predicted": -100, "true_label": "OPPOz"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSince December 10th, six people have lost their lives in avalanches. ⇔ ", "response": "Since December 10th six people have lost their lives in avalanches.", "predicted": 10, "true_label": "Humans died in an avalanche."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHumans died in an avalanche. ⇔ ", "response": "An avalanche has struck a popular skiing resort in Austria, killing at least 11 people.", "predicted": 11, "true_label": "Since December 10th, six people have lost their lives in avalanches."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在哪边呢 ⇔ ", "response": "你的意思是多少灵？求指教！", "predicted": -100, "true_label": "我爸那里"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我爸那里 ⇔ ", "response": "帮我爸爸", "predicted": -100, "true_label": "在哪边呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy is skillfully playing a piano ⇔ ", "response": "A boy is playing a piano", "predicted": -100, "true_label": "A boy is playing a piano"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA boy is playing a piano ⇔ ", "response": "A boy is playing a keyboard", "predicted": -100, "true_label": "A boy is skillfully playing a piano"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor7s ⇔ ", "response": "OPPOR7", "predicted": 7, "true_label": "oppor9s"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppor9s ⇔ ", "response": "OPPOR9", "predicted": 9, "true_label": "oppor7s"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n把QQ隐藏起来 ⇔ ", "response": "查看QQ消息", "predicted": -100, "true_label": "重新登录QQ"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n重新登录QQ ⇔ ", "response": "点击QQ登录", "predicted": -100, "true_label": "把QQ隐藏起来"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今年的冬天特别冷，尤其是北方地区。 ⇔ ", "response": "北方的冬天一直都很冷。", "predicted": -100, "true_label": "今年的冬天特别冷，尤其是北方地去。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今年的冬天特别冷，尤其是北方地去。 ⇔ ", "response": "今天早上我去了公园散步。", "predicted": -100, "true_label": "今年的冬天特别冷，尤其是北方地区。"}

eval_gen/gpu_results/results_gpu_6.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo dogs run together near the leaves. ⇔ ", "response": "Two dogs run near some leaves.", "predicted": -100, "true_label": "Two dogs are running."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo dogs are running. ⇔ ", "response": "Two dogsmoving rapidly in a circular motion.", "predicted": -100, "true_label": "Two dogs run together near the leaves."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nInvestigators uncovered a 4-inch bone fragment from beneath the concrete slab Thursday , but it turned out to be an animal bone , authorities said . ⇔ ", "response": "Investigators uncovered a 4-inch bone fragment from beneath a concrete slab last week , but it turned out to be an animal bone , authorities said on Tuesday .", "predicted": 4, "true_label": "Investigators uncovered a 4-inch bone fragment Thursday night , but authorities said it was from an animal ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nInvestigators uncovered a 4-inch bone fragment Thursday night , but authorities said it was from an animal . ⇔ ", "response": "Investigators uncovered a 4-inch bone fragment about 8:30 p.m . Thursday , but authorities said it was from an animal .", "predicted": 4, "true_label": "Investigators uncovered a 4-inch bone fragment from beneath the concrete slab Thursday , but it turned out to be an animal bone , authorities said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe ROE printed here were issued by General Jean Cot, then commander of U.N. forces, and were intended to establish the conditions under which the forces could use their weapons as they carry out the U.N. peacekeeping mission in Bosnia. ⇔ ", "response": "Jean Cot heads the U.N. forces.", "predicted": -100, "true_label": "U.N. peacekeeping forces withdrew from Bosnia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nU.N. peacekeeping forces withdrew from Bosnia. ⇔ ", "response": "The decision by the U.N. Security Council to end the peacekeeping force in Bosnia comes as a surprise even to those most familiar with the crisis.", "predicted": -100, "true_label": "The ROE printed here were issued by General Jean Cot, then commander of U.N. forces, and were intended to establish the conditions under which the forces could use their weapons as they carry out the U.N. peacekeeping mission in Bosnia."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no person pouring olive oil into a pot on the stove ⇔ ", "response": "A person is pouring olive oil into a pot on the stove", "predicted": -100, "true_label": "A person is pouring olive oil into a pot on the stove"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is pouring olive oil into a pot on the stove ⇔ ", "response": "A person is pouring oil for cooking into a pot", "predicted": -100, "true_label": "There is no person pouring olive oil into a pot on the stove"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n求解这是什么动漫！ ⇔ ", "response": "这是什么动漫？", "predicted": -100, "true_label": "这是什么动漫，求解"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这是什么动漫，求解 ⇔ ", "response": "这是什么动漫？", "predicted": -100, "true_label": "求解这是什么动漫！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMicrosoft is, undeniably, a company with monopoly power. ⇔ ", "response": "Microsoft has been accused of creating a monopoly.", "predicted": -100, "true_label": "Microsoft holds a monopoly power."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMicrosoft holds a monopoly power. ⇔ ", "response": "Microsoft maintains their very successful operating system monopoly through the use of proprietary specifications as well as aggressive marketing.", "predicted": -100, "true_label": "Microsoft is, undeniably, a company with monopoly power."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n想说什么就说什么 ⇔ ", "response": "说些什么", "predicted": -100, "true_label": "说什么好呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n说什么好呢 ⇔ ", "response": "好说什么好", "predicted": -100, "true_label": "想说什么就说什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBlond woman overlooking Seattle Space Needle scene. ⇔ ", "response": "A tourist checking out Seattle.", "predicted": -100, "true_label": "Someone taking in the space needle view."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone taking in the space needle view. ⇔ ", "response": "A man looks down from the space needle looking out over Seattle.", "predicted": -100, "true_label": "Blond woman overlooking Seattle Space Needle scene."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA group of children divided into a maroon team and a blue team playing a soccer game. ⇔ ", "response": "A group of children are playing soccer.", "predicted": -100, "true_label": "Some children are playing soccer by a school."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome children are playing soccer by a school. ⇔ ", "response": "Several children are playing soccer.", "predicted": -100, "true_label": "A group of children divided into a maroon team and a blue team playing a soccer game."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nhdr ⇔ ", "response": "OPPOR", "predicted": -100, "true_label": "hdjdjdrhr-duiJ"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nhdjdjdrhr-duiJ ⇔ ", "response": "hjjddrhrdj", "predicted": -100, "true_label": "hdr"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIt indicates , Robert said , “ that terrorists really don ’ t care who they attack . ⇔ ", "response": "Robert paraphrased : \" The bombings indicate that terrorists really don ' t care who they attack . \"", "predicted": -100, "true_label": "\" It also indicates the terrorists really don 't care who they attack . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" It also indicates the terrorists really don 't care who they attack . \" ⇔ ", "response": "\" The fact that they ( al-Qaida ) are not concerned who they attack shows how dangerous the situation really is .", "predicted": -100, "true_label": "It indicates , Robert said , “ that terrorists really don ’ t care who they attack ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国有几个皇帝 ⇔ ", "response": "飞机有几个轮子", "predicted": -100, "true_label": "抗美援朝的纪念日是多少"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n抗美援朝的纪念日是多少 ⇔ ", "response": "抗美援朝纪念日是12月13日", "predicted": 12, "true_label": "中国有几个皇帝"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天去逛街买了一条裙子 ⇔ ", "response": "我昨天去逛街买了一条秋裙", "predicted": -100, "true_label": "我昨天去逛街买了一条群子"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我昨天去逛街买了一条群子 ⇔ ", "response": "我昨天去逛街买了一条秋裤", "predicted": -100, "true_label": "我昨天去逛街买了一条裙子"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n是的，可以帮您提交受理单，找回支付密码和清空安全保护问题，可以吗 ⇔ ", "response": "可以，可以帮助您取消自动续费，找回支付密码和清空安全保护问题吗", "predicted": -100, "true_label": "这边帮您操作找回登录密码、找回正支付密码、清空安保问题的，您看可以吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这边帮您操作找回登录密码、找回正支付密码、清空安保问题的，您看可以吗？ ⇔ ", "response": "您的安全登录密码是：您的支付密码是：您的安保问题是：", "predicted": -100, "true_label": "是的，可以帮您提交受理单，找回支付密码和清空安全保护问题，可以吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n120秒等于多少分 ⇔ ", "response": "120秒等于多少分钟", "predicted": 120, "true_label": "1公里等于多少米"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n1公里等于多少米 ⇔ ", "response": "1公里等于多少米", "predicted": 1, "true_label": "120秒等于多少分"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA blond little boy in an orange sweatshirt with red sleeves is using scissors to cut something. ⇔ ", "response": "A boy is cutting with a pair of scissors.", "predicted": -100, "true_label": "A little male has clothes on with a pair of scissors in his hands."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA little male has clothes on with a pair of scissors in his hands. ⇔ ", "response": "A boy with clothes on is using a pair of scissors.", "predicted": -100, "true_label": "A blond little boy in an orange sweatshirt with red sleeves is using scissors to cut something."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhatcan ⇔ ", "response": "WHATS", "predicted": -100, "true_label": "Whatdoes"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhatdoes ⇔ ", "response": "WHATS", "predicted": -100, "true_label": "Whatcan"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天去公园散步了，风景真美丽 ⇔ ", "response": "我今天去公园散步，风景真的很好", "predicted": -100, "true_label": "我今天天气散步了，风景真美丽"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气散步了，风景真美丽 ⇔ ", "response": "我今天去散步了，风景真美丽", "predicted": -100, "true_label": "我今天去公园散步了，风景真美丽"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA worker peers out from atop a building under construction. ⇔ ", "response": "A man is atop of a building.", "predicted": -100, "true_label": "A person is atop of a building."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is atop of a building. ⇔ ", "response": "A man wearing a black shirt and sitting belt is standing on top of a white building.", "predicted": -100, "true_label": "A worker peers out from atop a building under construction."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and woman are talking ⇔ ", "response": "A man and woman are conversing", "predicted": -100, "true_label": "A man and a woman are silent"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are silent ⇔ ", "response": "A man and a woman are standing together, but they are silent", "predicted": -100, "true_label": "A man and woman are talking"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n茶壶里煮饺子下一句 ⇔ ", "response": "饺子放进煮开的茶壶里", "predicted": -100, "true_label": "茶壶里煮饺子的下一句"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n茶壶里煮饺子的下一句 ⇔ ", "response": "饺子的下一句 what：茶壶里煮饺子——升官了", "predicted": -100, "true_label": "茶壶里煮饺子下一句"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n开一下导航 ⇔ ", "response": "给我导", "predicted": -100, "true_label": "开一下电筒"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n开一下电筒 ⇔ ", "response": "给我关上电筒", "predicted": -100, "true_label": "开一下导航"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHampton Township is a few miles northeast of Bay City , about 100 miles away . ⇔ ", "response": "The township is located in Franklin County , about 100 miles ( 160 km ) northeast of Bay City .", "predicted": 100, "true_label": "Hampton Township is located a few kilometers northeast of Bay City , near Michigan 's Thumb ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHampton Township is located a few kilometers northeast of Bay City , near Michigan 's Thumb . ⇔ ", "response": "The township is located about 3 miles ( 5.6 km ) northeast of Bay City , Michigan 's Thumb .", "predicted": 3, "true_label": "Hampton Township is a few miles northeast of Bay City , about 100 miles away ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n周三需要了解的10件事 ⇔ ", "response": "昨天需要了解的5件事", "predicted": 5, "true_label": "周四需要了解的10件事"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n周四需要了解的10件事 ⇔ ", "response": "欧盟峰会周四开始", "predicted": -100, "true_label": "周三需要了解的10件事"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n艾肯的研究发表在9月9日的“纽约时报”上。24期“美国医学会杂志”。 ⇔ ", "response": "艾肯的研究发表在“纽约时报”上，日期是9月9日。他发表在24期“美国医学会杂志”上。", "predicted": 9, "true_label": "这一发现发表在周三的“美国医学会杂志”(Journal Of The American Medical Association)上。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这一发现发表在周三的“美国医学会杂志”(Journal Of The American Medical Association)上。 ⇔ ", "response": "这项研究结果将在周三出版的“美国医学会杂志”中公布。", "predicted": -100, "true_label": "艾肯的研究发表在9月9日的“纽约时报”上。24期“美国医学会杂志”。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该物种分布在墨西哥湾和加勒比海;在巴西的大西洋。 ⇔ ", "response": "该物种分布在墨西哥湾和加勒比海;在巴西的大西洋。", "predicted": -100, "true_label": "该物种分布在墨西哥湾和加勒比地区，沿大西洋的巴西。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该物种分布在墨西哥湾和加勒比地区，沿大西洋的巴西。 ⇔ ", "response": "该物种分布在墨西哥湾和加勒比地区，沿大西洋的巴西。", "predicted": -100, "true_label": "该物种分布在墨西哥湾和加勒比海;在巴西的大西洋。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe earthquake was felt throughout northern Japan, including Hokkaido's largest city, Sapporo, and Kushiro, on the island's eastern shore, but no reports of serious damage have been received. ⇔ ", "response": "An earthquake occurred off the coast of Japan.", "predicted": -100, "true_label": "An earthquake occurred in Hokkaido."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn earthquake occurred in Hokkaido. ⇔ ", "response": "A great earthquake occurred at 19:50:08 (UTC) on Thursday, September 25, 2003. The magnitude 8.0 event has been located in the Hokkaido, Japan region.", "predicted": 19, "true_label": "The earthquake was felt throughout northern Japan, including Hokkaido's largest city, Sapporo, and Kushiro, on the island's eastern shore, but no reports of serious damage have been received."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe dog is playing with a plastic container ⇔ ", "response": "The puppy is playing with a plastic container", "predicted": -100, "true_label": "The puppy is not playing with a plastic container"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe puppy is not playing with a plastic container ⇔ ", "response": "The puppy is playing with a plastic container", "predicted": -100, "true_label": "The dog is playing with a plastic container"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n三个臭皮匠顶个诸葛亮的意思是什么 ⇔ ", "response": "三个臭皮匠，顶个诸葛亮，意思是什么", "predicted": -100, "true_label": "有关天气的谚语"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有关天气的谚语 ⇔ ", "response": "天气对体育比赛的影响", "predicted": -100, "true_label": "三个臭皮匠顶个诸葛亮的意思是什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nClosed sessions are routinely held at the United Nations tribunal that deals with Balkan war crimes , but usually to protect witnesses 's safety . ⇔ ", "response": "Closed sessions are held at the court 's Banff County Tribunal dealing with the Balkans , but for different reasons .", "predicted": -100, "true_label": "Closed sessions are routinely held at the U.N. tribunal that deals with Balkan war crimes , but they are usually closed to protect witnesses who fear for their safety ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nClosed sessions are routinely held at the U.N. tribunal that deals with Balkan war crimes , but they are usually closed to protect witnesses who fear for their safety . ⇔ ", "response": "They are usually closed to protect witnesses who fear for their safety .", "predicted": -100, "true_label": "Closed sessions are routinely held at the United Nations tribunal that deals with Balkan war crimes , but usually to protect witnesses 's safety ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n西游记里的人物 ⇔ ", "response": "西游记里有哪些人物", "predicted": -100, "true_label": "《西游记》里的人物"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n《西游记》里的人物 ⇔ ", "response": "以下是我喜欢的四本书及对应的作者：", "predicted": -100, "true_label": "西游记里的人物"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAxocomanitla是墨西哥特拉斯卡拉东南部的一个自治市。 ⇔ ", "response": "Axocomanitla是墨西哥特拉斯卡拉东南部的一个自治市。", "predicted": -100, "true_label": "San Lorenzo Axocomanitla是墨西哥东南部特拉斯卡拉的一个自治市。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSan Lorenzo Axocomanitla是墨西哥东南部特拉斯卡拉的一个自治市。 ⇔ ", "response": "San Lorenzo Axocomanitla是墨西哥东南部特拉斯卡拉的一个自治市。", "predicted": -100, "true_label": "Axocomanitla是墨西哥特拉斯卡拉东南部的一个自治市。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n1992年12月4日 - 剑桥联合训练师Ian Atkins被任命为伯明翰市教练。 ⇔ ", "response": "1992年12月4日，剑桥联合训练师Ian Atkins被任命为伯明翰市教练。", "predicted": 1992, "true_label": "1992年12月4日 - 剑桥联队主教练伊恩·阿特金斯被任命为伯明翰市的经理。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n1992年12月4日 - 剑桥联队主教练伊恩·阿特金斯被任命为伯明翰市的经理。 ⇔ ", "response": "1992年12月4日，伊恩·阿特金斯被任命为伯明翰市的经理。\n1992年12月4日，伊恩·阿特金斯被任命为伯明翰市的经理。", "predicted": 1992, "true_label": "1992年12月4日 - 剑桥联合训练师Ian Atkins被任命为伯明翰市教练。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这么看来，白金汉就是拉罗谢尔人的最后一线希望，就是他们的救世主了。 ⇔ ", "response": "这么看来，他便是拉罗谢尔人的最后一线希望，他们的救世主了。", "predicted": -100, "true_label": "白金汉是他们的救世主。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n白金汉是他们的救世主。 ⇔ ", "response": "白金汉是他们的救星。", "predicted": -100, "true_label": "这么看来，白金汉就是拉罗谢尔人的最后一线希望，就是他们的救世主了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nChichigalpa的土着人是托尔特克人，Niquiranos和Chorotegas。 ⇔ ", "response": "Chichigalpa的土着人是托尔特克人，Niquiranos和Chorotegas。", "predicted": -100, "true_label": "源自土生土长的Chichigalpa的Toltec人是Niquiranos和Chorotegas。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n源自土生土长的Chichigalpa的Toltec人是Niquiranos和Chorotegas。 ⇔ ", "response": "Toltec人源自土生土长的Chichigalpa的Niquiranos和Chorotegas。", "predicted": -100, "true_label": "Chichigalpa的土着人是托尔特克人，Niquiranos和Chorotegas。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWeintraub发现了彼得，保罗和玛丽，Randy Newman（他因淫秽而被捕），Lenny Bruce和The Isley Brothers等行为。 ⇔ ", "response": "Weintraub“发现了”彼得，保罗和玛丽，Randy Newman（他因淫秽而被捕），Lenny Bruce和The Isley Brothers等艺人。", "predicted": -100, "true_label": "葡萄发现了彼得，保罗和玛丽，兰迪纽曼（他因淫秽而被捕），莱尼布鲁斯和伊斯利兄弟等行为。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n葡萄发现了彼得，保罗和玛丽，兰迪纽曼（他因淫秽而被捕），莱尼布鲁斯和伊斯利兄弟等行为。 ⇔ ", "response": "葡萄发现彼得，保罗和玛丽，兰迪纽曼（他因淫秽而被捕），莱尼布鲁斯和伊斯利兄弟等不洁行为。", "predicted": -100, "true_label": "Weintraub发现了彼得，保罗和玛丽，Randy Newman（他因淫秽而被捕），Lenny Bruce和The Isley Brothers等行为。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child is walking down the street and a jeep is pulling up ⇔ ", "response": "A child is walking down the street and a car is pulling up", "predicted": -100, "true_label": "A child is walking down the street and a jeep is stopping"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child is walking down the street and a jeep is stopping ⇔ ", "response": "A child is walking down the street and the driver of the jeep is looking in the window", "predicted": -100, "true_label": "A child is walking down the street and a jeep is pulling up"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么本机暂时不支持镜像投屏 ⇔ ", "response": "为什么我的镜像不能投屏", "predicted": -100, "true_label": "手机不支持镜像投屏怎么办"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n手机不支持镜像投屏怎么办 ⇔ ", "response": "手机投屏显示很小怎么办", "predicted": -100, "true_label": "为什么本机暂时不支持镜像投屏"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n数百名荣誉学生在俄亥俄枪击案中丧生 ⇔ ", "response": "几名学生在一所学校的枪击案中丧生", "predicted": -100, "true_label": "俄亥俄州枪击案受害者的葬礼"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n俄亥俄州枪击案受害者的葬礼 ⇔ ", "response": "1962年佛罗里达洲际公路车撞人死亡六人", "predicted": 1962, "true_label": "数百名荣誉学生在俄亥俄枪击案中丧生"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n迪达勒斯先生从上衣下摆底下掏出烟草袋和烟斗。 ⇔ ", "response": "迪达勒斯先生从衣摆下拿出一个烟草袋和一个烟斗。", "predicted": -100, "true_label": "代达勒斯先生从上衣口袋里掏出了烟袋、烟斗。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n代达勒斯先生从上衣口袋里掏出了烟袋、烟斗。 ⇔ ", "response": "代达勒斯先生从上衣口袋里掏出一根烟袋管。", "predicted": -100, "true_label": "迪达勒斯先生从上衣下摆底下掏出烟草袋和烟斗。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n区内阁由Anne Marit Mevassvik领导，有四名成员，来自保守党，工党和基督教民主党。 ⇔ ", "response": "区内阁由Anne Marit Mevassvik领导，有四名成员，来自保守党，工党和基督教民主党。", "predicted": -100, "true_label": "县内阁由Anne Marit Mevassvik领导，有四名成员，分别来自工党，保守党和基督教民主党。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n县内阁由Anne Marit Mevassvik领导，有四名成员，分别来自工党，保守党和基督教民主党。 ⇔ ", "response": "县内阁由Anne Marit Mevassvik领导，有四名成员，分别来自工党，保守党和基督教民主党。", "predicted": -100, "true_label": "区内��由Anne Marit Mevassvik领导，有四名成员，来自保守党，工党和基督教民主党。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n刚刚存储个别图片删除恢复 ⇔ ", "response": "存储个别图片删除恢复", "predicted": -100, "true_label": "恢复刚刚删的所有图片"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n恢复刚刚删的所有图片 ⇔ ", "response": "打开微信，点击“相册”", "predicted": -100, "true_label": "刚刚存储个别图片删除恢复"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n“杰里米是个好人，”巴伯说，并补充道：“杰里米正过着这位纽约运动员梦寐以求的生活。” ⇔ ", "response": "“杰里米是个好人，”巴伯说。并补充说，杰里米正过着这位纽约运动员梦寐以求的生活。", "predicted": -100, "true_label": "他还说沙基“过着纽约运动员梦寐以求的生活。”"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n他还说沙基“过着纽约运动员梦寐以求的生活。” ⇔ ", "response": "沙基还告诉他们“过着纽约运动员梦寐以求的生活”。", "predicted": -100, "true_label": "“杰里米是个好人，”巴伯说，并补充道：“杰里米正过着这位纽约运动员梦寐以求的生活。”"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么夏天下雨打雷，冬天下雪不打雷？ ⇔ ", "response": "为什么夏天下雨打雷，冬天下雪不打雷？", "predicted": -100, "true_label": "为什么夏天下雨打雷，冬天下雪就不打雷呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么夏天下雨打雷，冬天下雪就不打雷呢？ ⇔ ", "response": "为什么冬天下雪时不打雷，夏天下雨时却打雷", "predicted": -100, "true_label": "为什么夏天下雨打雷，冬天下雪不打雷？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这个是一个无效的文本 ⇔ ", "response": "这是一条有效的文本", "predicted": -100, "true_label": "今天的天气不错"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天的天气不错 ⇔ ", "response": "我喜欢阅读科幻小说", "predicted": -100, "true_label": "这个是一个无效的文本"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA golfer is getting ready to putt on the green, with a crowd of people watching in the background. ⇔ ", "response": "A golfer is putting on a green, with a crowd of people watching.", "predicted": -100, "true_label": "A golfer readies to putt the ball."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA golfer readies to putt the ball. ⇔ ", "response": "A golfer prepares to complete his round by putting the ball into the hole.", "predicted": -100, "true_label": "A golfer is getting ready to putt on the green, with a crowd of people watching in the background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBuoyed by some of the advice imparted by Nicklaus , Howell shot an 8-under 64 for a one-stroke lead over Kenny Perry . ⇔ ", "response": "Howell also benefited from advice from Nicklaus , who shot an 8-under 64 for a one-stroke lead over Kenny Perry .", "predicted": 8, "true_label": "Buoyed by advice imparted by Nicklaus , Howell shot an 8-under 64 on Thursday to enter today 's round with a one-stroke lead over Kenny Perry ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBuoyed by advice imparted by Nicklaus , Howell shot an 8-under 64 on Thursday to enter today 's round with a one-stroke lead over Kenny Perry . ⇔ ", "response": "Howell got as many as two strokes on the competition with an 8-under 64 on Thursday .", "predicted": 8, "true_label": "Buoyed by some of the advice imparted by Nicklaus , Howell shot an 8-under 64 for a one-stroke lead over Kenny Perry ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nchinaunicomvol ⇔ ", "response": "volareinchina", "predicted": -100, "true_label": "mongolia"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nmongolia ⇔ ", "response": "n06", "predicted": 6, "true_label": "chinaunicomvol"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRELAX ⇔ ", "response": "Watch", "predicted": -100, "true_label": "RelaxING"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRelaxING ⇔ ", "response": "Enjoyinglife", "predicted": -100, "true_label": "RELAX"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOPPOA79t ⇔ ", "response": "oppoa79tota", "predicted": 79, "true_label": "oppoa79ota"}
+{"prompt": "Please generate a paraphrase for the following sentence.\noppoa79ota ⇔ ", "response": "OPPOa4", "predicted": 4, "true_label": "OPPOA79t"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n怎么没有声音了 ⇔ ", "response": "怎么没有声了", "predicted": -100, "true_label": "有什么游戏可以玩吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有什么游戏可以玩吗 ⇔ ", "response": "有那个王者的游戏可以玩吗", "predicted": -100, "true_label": "怎么没有声音了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个女人和一个男人被耳机逗乐了。 ⇔ ", "response": "一男一女在舞台上表演。", "predicted": -100, "true_label": "一男一女运动后休息。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一男一女运动后休息。 ⇔ ", "response": "一个男人和一个女人正在吃面。", "predicted": -100, "true_label": "一个女人和一个男人被耳机逗乐了。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a red shirt and a blue hat is sitting on a sidewalk and touching drinking glasses. ⇔ ", "response": "The man is touching drinks.", "predicted": -100, "true_label": "a guy with a hat sits with a glass"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na guy with a hat sits with a glass ⇔ ", "response": "A man sips alcohol from a glass while wearing a baseball cap.", "predicted": -100, "true_label": "A man in a red shirt and a blue hat is sitting on a sidewalk and touching drinking glasses."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe incubation period in cattle is four to five years , said Stephen Sundlof of the U.S. Food and Drug Administration . ⇔ ", "response": "\" The incubation period in cattle is four to five years , \" Sundlof said .", "predicted": -100, "true_label": "The incubation period in cattle is four to five years , said Dr. Stephen Sundlof of the Food and Drug Administration ( news - web sites ) ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe incubation period in cattle is four to five years , said Dr. Stephen Sundlof of the Food and Drug Administration ( news - web sites ) . ⇔ ", "response": "The incubation period is four to five years , according to Dr. Stephen Sundlof of the Food and Drug Administration .", "predicted": -100, "true_label": "The incubation period in cattle is four to five years , said Stephen Sundlof of the U.S. Food and Drug Administration ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nYesIam ⇔ ", "response": "Iama", "predicted": -100, "true_label": "Yesyesyes"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nYesyesyes ⇔ ", "response": "Yesyes", "predicted": -100, "true_label": "YesIam"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n袼劳库斯转身对在座的人说:出卖我的就是这个人，他害得我家破人亡基督教徒们全都知道他的不幸遭遇，维尼茨尤斯也听说过他，不过维尼茨尤斯没有想到他就是格劳库斯，因为这位医生刚才给他包扎伤口时，他已经痛得晕过去了，没有听见他的名字。 ⇔ ", "response": "袼劳库斯转身对说：“就是这个人出卖了我，我家的遭遇，基督教徒们都很知道，维尼茨尤斯也听说过我，可是他没想到我就是格劳库斯。”", "predicted": -100, "true_label": "对戈劳库斯的家史，这里的每个人都非常清楚，只是维尼裘斯不曾想到戈劳库斯就是为他包扎伤口的人，因为当时，他疼得昏死过去，并没有听到这个人的名字。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n对戈劳库斯的家史，这里的每个人都非常清楚，只是维尼裘斯不曾想到戈劳库斯就是为他包扎伤口的人，因为当时，他疼得昏死过去，并没有听到这个人的名字。 ⇔ ", "response": "戈劳库斯可是我们家族的人，可是维尼裘斯可没这个想法，当时他受的伤很重，几乎要昏过去，包扎他的就是那个姓戈劳库斯的人，可是他名字却叫不出来，听不出来。", "predicted": -100, "true_label": "袼劳库斯转身对在座的人说:出卖我的就是这个人，他害得我家破人亡基督教徒们全都知道他的不幸遭遇，维尼茨尤斯也听说过他，不过维尼茨尤斯没有想到他就是格劳库斯，因为这位医生刚才给他包扎伤口时，他已经痛得晕过去了，没有听见他的名字。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n法国的首都在哪里 ⇔ ", "response": "法国，首都在哪里？", "predicted": -100, "true_label": "法国的首都是哪里？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n法国的首都是哪里？ ⇔ ", "response": "法国哪个城市有科技展", "predicted": -100, "true_label": "法国的首都在哪里"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n鳄鱼害怕什么颜色 ⇔ ", "response": "鳄鱼什么颜色", "predicted": -100, "true_label": "鳄鱼害怕什么颜色？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n鳄鱼害怕什么颜色？ ⇔ ", "response": "鳄鱼害怕什么食物", "predicted": -100, "true_label": "鳄鱼害怕什么颜色"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n发短信给老婆 ⇔ ", "response": "给爸爸发短信", "predicted": -100, "true_label": "请打开短信"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n请打开短信 ⇔ ", "response": "发送短信", "predicted": -100, "true_label": "发短信给老婆"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我要听上海话 ⇔ ", "response": "我要听英语", "predicted": -100, "true_label": "打开放烟花的声音"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n打开放烟花的声音 ⇔ ", "response": "观看烟花是什么感觉", "predicted": -100, "true_label": "我要听上海话"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n突触簇形成是指将树突棘添加到通过先前学习添加了其他刺的新区域。 ⇔ ", "response": "区域通过将树突棘添加到先前已经添加其他刺的新区域来形成突触簇。", "predicted": -100, "true_label": "突触聚类是指将树突棘添加到通过先前学习添加了其他刺的新区域。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n突触聚类是指将树突棘添加到通过先前学习添加了其他刺的新区域。 ⇔ ", "response": "突触减少是指将树突棘从通过先前学习删除其他刺的区域添加到的新区域。", "predicted": -100, "true_label": "突触簇形成是指将树突棘添加到通过先前学习添加了其他刺的新区域。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe soccer team, clad in blue for the match, began to counter down the field in front of the defender, clad in red. ⇔ ", "response": "A soccer game is taking place between a team wearing blue and red.", "predicted": -100, "true_label": "The soccer team in blue plays soccer."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe soccer team in blue plays soccer. ⇔ ", "response": "A team is playing.", "predicted": -100, "true_label": "The soccer team, clad in blue for the match, began to counter down the field in front of the defender, clad in red."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个穿着灰色西装的男人跪在一个穿着白色连衣裙的女人面前。 ⇔ ", "response": "一名男子在婚礼上跪下，向穿着黑色礼服的女人求婚。", "predicted": -100, "true_label": "一个穿红衣服的女人和一杯啤酒，后面跟着一个穿白色衣服的男人。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个穿红衣服的女人和一杯啤酒，后面跟着一个穿白色衣服的男人。 ⇔ ", "response": "一个穿红衣服的女人正握住一杯啤酒，背后是一个穿白色衬衫的男人。", "predicted": -100, "true_label": "一个穿着灰色西装的男人跪在一个穿着白色连衣裙的女人面前。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天我们去动物园玩 ⇔ ", "response": "明天我们去玩动物园", "predicted": -100, "true_label": "明天我们去动玩物园"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天我们去动玩物园 ⇔ ", "response": "明天我们去动园玩", "predicted": -100, "true_label": "明天我们去动物园玩"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了午餐 ⇔ ", "response": "我今天天吃了午餐", "predicted": -100, "true_label": "我今天吃了误餐"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了误餐 ⇔ ", "response": "我今天吃了饭", "predicted": -100, "true_label": "我今天吃了午餐"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA clown is singing and people are dancing ⇔ ", "response": "The man is singing and the people are dancing", "predicted": -100, "true_label": "A costumed performer is singing and people are dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA costumed performer is singing and people are dancing ⇔ ", "response": "A costumed performer is singing and nobody is dancing", "predicted": -100, "true_label": "A clown is singing and people are dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你知道什么游戏 ⇔ ", "response": "你玩什么游戏", "predicted": -100, "true_label": "我的游戏呢"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的游戏呢 ⇔ ", "response": "游戏出理啊", "predicted": -100, "true_label": "你知道什么游戏"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n与日俱增 ⇔ ", "response": "日益增加", "predicted": -100, "true_label": "自作自受的近义词"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n自作自受的近义词 ⇔ ", "response": "自作自受的反义词", "predicted": -100, "true_label": "与日俱增"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n波多黎各印加省是秘鲁Huánuco地区11个省中最大的省份。 ⇔ ", "response": "波多黎各印加省是秘鲁Huánuco地区11个省中最大的省份。", "predicted": 11, "true_label": "Puerto Inca省是秘鲁Huánuco地区11个省中最大的省份。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPuerto Inca省是秘鲁Huánuco地区11个省中最大的省份。 ⇔ ", "response": "Puerto Inca省是秘鲁胡纳努科地区11个省中最大的省份。", "predicted": 11, "true_label": "波多黎各印加省是秘鲁Huánuco地区11个省中最大的省份。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n24的因数有 ⇔ ", "response": "24的因数有", "predicted": 24, "true_label": "等腰三角形的周长怎么算"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n等腰三角形的周长怎么算 ⇔ ", "response": "等腰三角形的周长怎么求", "predicted": -100, "true_label": "24的因数有"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBetty ⇔ ", "response": "GEORGE", "predicted": -100, "true_label": "PRETTY"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPRETTY ⇔ ", "response": "BEAUTIFUL", "predicted": -100, "true_label": "Betty"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young woman is shaking hands with a man ⇔ ", "response": "A woman is shaking hands with a man", "predicted": -100, "true_label": "A man and a woman are shaking hands"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a woman are shaking hands ⇔ ", "response": "A man and a woman are standing near each other and shaking hands", "predicted": -100, "true_label": "A young woman is shaking hands with a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is eating some food ⇔ ", "response": "There is no man eating food", "predicted": -100, "true_label": "The man is fasting"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is fasting ⇔ ", "response": "The man is fasting for Islam", "predicted": -100, "true_label": "A man is eating some food"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA female softball player wearing blue and red crouches in the infield, waiting for the next play. ⇔ ", "response": "A softball player is playing.", "predicted": -100, "true_label": "a player is waiting"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na player is waiting ⇔ ", "response": "A soccer player waits on the bench for the game to start.", "predicted": -100, "true_label": "A female softball player wearing blue and red crouches in the infield, waiting for the next play."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe Supreme Court long ago held that students could not be compelled to join in the pledge . ⇔ ", "response": "Last year , the Supreme Court held that students could challenge the pledge , known as the Pledge of Allegiance , in school .", "predicted": -100, "true_label": "The U.S. Supreme Court has previously ruled that students are not compelled to say the Pledge of Allegiance ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe U.S. Supreme Court has previously ruled that students are not compelled to say the Pledge of Allegiance . ⇔ ", "response": "Students currently are not required to say the Pledge of Allegiance in school , the <i> New York Times </i> reported .", "predicted": -100, "true_label": "The Supreme Court long ago held that students could not be compelled to join in the pledge ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWHATSYOURFATHER ⇔ ", "response": "Whatsyour", "predicted": -100, "true_label": "Whatsyourhobby"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWhatsyourhobby ⇔ ", "response": "HOLA", "predicted": -100, "true_label": "WHATSYOURFATHER"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nsn是什么呀你直接说不就行了 ⇔ ", "response": "你看这个题，直接说不就行了，还要说什么sn是什么呀", "predicted": -100, "true_label": "要是不说我就认为你是不男不女"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n要是不说我就认为你是不男不女 ⇔ ", "response": "你都是女的，怎么不是女的呢", "predicted": -100, "true_label": "sn是什么呀你直接说不就行了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man in a suit standing at a microphone and singing ⇔ ", "response": "A man in a suit is standing at a microphone and singing", "predicted": -100, "true_label": "A man in a suit is standing at a microphone and singing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a suit is standing at a microphone and singing ⇔ ", "response": "The man on stage is singing into the microphone", "predicted": -100, "true_label": "There is no man in a suit standing at a microphone and singing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man wearing a short-sleeved blue shirt and carrying a blue backpack while using snow walking sticks treks through the snow with a woman wearing a long-sleeved blue shirt and black pants also using snow walking sticks. ⇔ ", "response": "A man and a woman use snow walking sticks.", "predicted": -100, "true_label": "a guy with a blue top carries a pack and makes his way through snow"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na guy with a blue top carries a pack and makes his way through snow ⇔ ", "response": "A man in a blue jacket is carrying a backpack and walking in the snow.", "predicted": -100, "true_label": "A man wearing a short-sleeved blue shirt and carrying a blue backpack while using snow walking sticks treks through the snow with a woman wearing a long-sleeved blue shirt and black pants also using snow walking sticks."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n男人带什么项链好 ⇔ ", "response": "带什么项链好", "predicted": -100, "true_label": "男人带什么玉项链好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n男人带什么玉项链好 ⇔ ", "response": "送什么玉项链比较好", "predicted": -100, "true_label": "男人带什么项链好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAfter Saddam 's regime crumbled in early April , it had to wait for legal hurdles to be crossed before sales could resume . ⇔ ", "response": "It had to wait for legal hurdles to be crossed before sales could resume 16 days after Saddam Hussein 's regime crumbled .", "predicted": 16, "true_label": "After Saddam 's regime crumbled in early April , legal hurdles had to be cleared before sales could resume ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAfter Saddam 's regime crumbled in early April , legal hurdles had to be cleared before sales could resume . ⇔ ", "response": "The oil sales to Iran , Iraq 's former ally , were blocked after Saddam Hussein 's regime collapsed in early April .", "predicted": -100, "true_label": "After Saddam 's regime crumbled in early April , it had to wait for legal hurdles to be crossed before sales could resume ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喜欢在下半场跑比第一节快1-2分钟。 ⇔ ", "response": "我在第一堂课只跑了1-2分钟。", "predicted": 1, "true_label": "我肯定会在上半场稍微慢一点。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我肯定会在上半场稍微慢一点。 ⇔ ", "response": "上半场我会稍微慢一点。", "predicted": -100, "true_label": "我喜欢在下半场跑比第一节快1-2分钟。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n孟买是印度第二大生活成本最高的城市。 ⇔ ", "response": "据信，孟买是印度生活成本最高的城市。", "predicted": -100, "true_label": "印度科技之都班加罗尔是印度生活成本最高的城市：调查"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n印度科技之都班加罗尔是印度生活成本最高的城市：调查 ⇔ ", "response": "印度生活成本最高的城市：调查", "predicted": -100, "true_label": "孟买是印度第二大生活成本最高的城市。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCHINESE ⇔ ", "response": "Neighbor", "predicted": -100, "true_label": "Chinesebook"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nChinesebook ⇔ ", "response": "bookcase", "predicted": -100, "true_label": "CHINESE"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你喜欢范丞丞吗 ⇔ ", "response": "你喜欢丞丞吗", "predicted": -100, "true_label": "王者荣耀有多少个英雄"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n王者荣耀有多少个英雄 ⇔ ", "response": "王者荣耀有多少英雄", "predicted": -100, "true_label": "你喜欢范丞丞吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is not kick boxing with a trainer ⇔ ", "response": "The man is kick boxing with a trainer", "predicted": -100, "true_label": "The man is kick boxing with a trainer"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is kick boxing with a trainer ⇔ ", "response": "The man is kicking a boxing trainer", "predicted": -100, "true_label": "The man is not kick boxing with a trainer"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么东西越热它爬的越高。 ⇔ ", "response": "猫喜欢爬高处，热东���", "predicted": -100, "true_label": "什么东西越热爬的越高"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么东西越热爬的越高 ⇔ ", "response": "东西越冷缩的越高", "predicted": -100, "true_label": "什么东西越热它爬的越高。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你会说什么 ⇔ ", "response": "你说我说你怎么样", "predicted": -100, "true_label": "你会说白话吗"}

eval_gen/gpu_results/results_gpu_7.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please generate a paraphrase for the following sentence.\n你会说白话吗 ⇔ ", "response": "你会说粤语吗", "predicted": -100, "true_label": "你会说什么"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n埃及逮捕穆斯林兄弟会领导人巴迪 ⇔ ", "response": "埃及逮捕穆斯林兄弟会领导人", "predicted": -100, "true_label": "埃及准备应对更多的暴力，穆斯林兄弟会称其为“愤怒的游行”。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n埃及准备应对更多的暴力，穆斯林兄弟会称其为“愤怒的游行”。 ⇔ ", "response": "埃及总理表示，如果再次出现抗议活动，该国将准备应对更多的暴力。", "predicted": -100, "true_label": "埃及逮捕穆斯林兄弟会领导人巴迪"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThey came despite what BA called a \" difficult quarter \" , which it said included unofficial industrial action at Heathrow . ⇔ ", "response": "Passengers arrived in London tonight despite ba calling of a \" difficult quarter \" which it said included unofficial industrial action at heathrow .", "predicted": -100, "true_label": "BA said the second quarter , which included unofficial industrial action at Heathrow , had been difficult ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nBA said the second quarter , which included unofficial industrial action at Heathrow , had been difficult . ⇔ ", "response": "The second quarter included unofficial industrial action at Heathrow , and had been difficult .", "predicted": -100, "true_label": "They came despite what BA called a \" difficult quarter \" , which it said included unofficial industrial action at Heathrow ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA woman is holding her child while a man is reading something from a piece of paper and taking something from a bowl that a younger girl is holding in a park with a small lake in the background. ⇔ ", "response": "A man reads while a woman holds her child.", "predicted": -100, "true_label": "There are people outside"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere are people outside ⇔ ", "response": "A few people are walking in the rain.", "predicted": -100, "true_label": "A woman is holding her child while a man is reading something from a piece of paper and taking something from a bowl that a younger girl is holding in a park with a small lake in the background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了午饭 ⇔ ", "response": "我今天天气午饭", "predicted": -100, "true_label": "我今天天气吃了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天天气吃了 ⇔ ", "response": "我今天吃了饭", "predicted": -100, "true_label": "我今天吃了午饭"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nNickerson、Duquette、Bruno、Askov、Sandstone 和 Sturgeon Lake 社区都在克里克附近。 ⇔ ", "response": "Nickerson、Duquette、Bruno、Askov、Sandstone 和 Sturgeon Lake 社区都在克里克以南。", "predicted": -100, "true_label": "布鲁诺、度凯特、尼克森、阿斯科夫、桑德斯东和特金莱克社区都离克里克很近。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n布鲁诺、度凯特、尼克森、阿斯科夫、桑德斯东和特金莱克社区都离克里克很近。 ⇔ ", "response": "布鲁诺、度凯特、尼克森、阿斯科夫、桑德斯东和特金莱克社区都离克里克很近。", "predicted": -100, "true_label": "Nickerson、Duquette、Bruno、Askov、Sandstone 和 Sturgeon Lake 社区都在克里克附近。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe Longuet and Leotard Affairs (1996) concern both personal privileges for, and illegal financing of, the Republican Party leaders. ⇔ ", "response": "The Longuet and Leotard Party leaders were illegally financed.", "predicted": -100, "true_label": "A French political party was illegally financed."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA French political party was illegally financed. ⇔ ", "response": "The money came from Xenophon Tukyulagi, a French lawstudent who has become embroiled with several Frenchpolitical parties as an activist.", "predicted": -100, "true_label": "The Longuet and Leotard Affairs (1996) concern both personal privileges for, and illegal financing of, the Republican Party leaders."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe rhino is grazing on the grass ⇔ ", "response": "A rhino is moving around a enclosure", "predicted": -100, "true_label": "A animal is grazing in a field"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA animal is grazing in a field ⇔ ", "response": "A horse is grazing in a field", "predicted": -100, "true_label": "The rhino is grazing on the grass"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么牌子的净水器好 ⇔ ", "response": "净水器什么牌子好", "predicted": -100, "true_label": "什么牌子净水器好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么牌子净水器好 ⇔ ", "response": "净水器什么牌子好", "predicted": -100, "true_label": "什么牌子的净水器好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n查询别人手机话费余额 ⇔ ", "response": "手机话费查询余额", "predicted": -100, "true_label": "如何在网上查询别人手机话费余额"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n如何在网上查询别人手机话费余额 ⇔ ", "response": "如何查询手机话费余额", "predicted": -100, "true_label": "查询别人手机话费余额"}
+{"prompt": "Please generate a paraphrase for the following sentence.\npccm 00 ⇔ ", "response": "PCCM00", "predicted": 0, "true_label": "OTC"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOTC ⇔ ", "response": "ANYTHING", "predicted": -100, "true_label": "pccm 00"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有什么好看的恐怖电影吗？ ⇔ ", "response": "你觉得《恐怖电影》怎么样？", "predicted": -100, "true_label": "有什么好看的恐怖电影吗?"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n有什么好看的恐怖电影吗? ⇔ ", "response": "有一些好看的恐怖电影吗？", "predicted": -100, "true_label": "有什么好看的恐怖电影吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWorkers standing on a lift. ⇔ ", "response": "Lift with several workers inside", "predicted": -100, "true_label": "Workers stand on a lift"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nWorkers stand on a lift ⇔ ", "response": "Workers are lifting a lift", "predicted": -100, "true_label": "Workers standing on a lift."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n二十分钟倒计时开始是几时 ⇔ ", "response": "倒计时几时开始", "predicted": -100, "true_label": "我想要二十分钟提醒我"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我想要二十分钟提醒我 ⇔ ", "response": "我要定二十点的提醒", "predicted": -100, "true_label": "二十分钟倒计时开始是几时"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA busy street with numerous people interacting and going about their lives. ⇔ ", "response": "A street is crowded with people.", "predicted": -100, "true_label": "The street is alive with activity."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe street is alive with activity. ⇔ ", "response": "A busy street with shops and people.", "predicted": -100, "true_label": "A busy street with numerous people interacting and going about their lives."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man with a tan jacket with a full grocery bag is crossing the street. ⇔ ", "response": "A man is crossing the street outside of the crosswalk.", "predicted": -100, "true_label": "The man just went shopping at the grocery store."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man just went shopping at the grocery store. ⇔ ", "response": "A man in a black jacket, blue jeans and a striped shirt brings a grocery bag to a plastic mailbox and pushes a stroller in the opposite direction.", "predicted": -100, "true_label": "A man with a tan jacket with a full grocery bag is crossing the street."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nScp96 ⇔ ", "response": "scp9696", "predicted": 9696, "true_label": "scp001"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nscp001 ⇔ ", "response": "SCP003", "predicted": 3, "true_label": "Scp96"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n两个村庄位于波蒂奇（Portage）：南部的杰里市（Jerry City）和西北部的波蒂奇（Portage）镇的一部分。 ⇔ ", "response": "两个村庄位于波蒂奇：南部的杰里市和西北部Portage镇。", "predicted": -100, "true_label": "在波蒂奇镇有两个村庄：南部的杰瑞城的一部分和西北部的波蒂奇的一部分。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在波蒂奇镇有两个村庄：南部的杰瑞城的一部分和西北部的波蒂奇的一部分。 ⇔ ", "response": "波特奇镇有两个村庄：南部杰瑞城的一部分和西北部分。", "predicted": -100, "true_label": "两个村庄位于波蒂奇（Portage）：南部的杰里市（Jerry City）和西北部的波蒂奇（Portage）镇的一部分。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n然而，格鲁吉亚语在指定数量时使用单数形式，因此在实践中，“tetri”的复数只是“tetri”。 ⇔ ", "response": "然而，格鲁吉亚语在指定数量时使用单数形式，因此在实践中，“tetri”的复数只是“tetri”。", "predicted": -100, "true_label": "然而，当指定数量时，格鲁吉亚语言是单数形式，因此在实践中，“tetri”的复数仅使用“tetri”。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n然而，当指定数量时，格鲁吉亚语言是单数形式，因此在实践中，“tetri”的复数仅使用“tetri”。 ⇔ ", "response": "然而，当指定数量时，格鲁吉亚语言是单数形式，因此在实践中，“tetri”的复数仅使用“tetri”。", "predicted": -100, "true_label": "然而，格鲁吉亚语在指定数量时使用单数形式，因此在实践中，“tetri”的复数只是“tetri”。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n开船的时候，头里就觉得极不舒服。 ⇔ ", "response": "在海上航行的日子，头里就觉得极不舒服。", "predicted": -100, "true_label": "开船的时候，他就感到头疼。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n开船的时候，他就感到头疼。 ⇔ ", "response": "开了很长时间的车，他感到头昏眼花。", "predicted": -100, "true_label": "开船的时候，头里就觉得极不舒服。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么星座的女生最漂亮 ⇔ ", "response": "哪个星座最性感", "predicted": -100, "true_label": "什么星座的女生最漂亮？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n什么星座的女生最漂亮？ ⇔ ", "response": "哪个星座最漂亮", "predicted": -100, "true_label": "什么星座的女生最漂亮"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我喝酸奶你舔酸奶盖盖 ⇔ ", "response": "我不喝酸奶你舔酸奶盖子", "predicted": -100, "true_label": "西红柿炒鸡蛋的做法"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n西红柿炒鸡蛋的做法 ⇔ ", "response": "鸡蛋西红柿怎么吃？", "predicted": -100, "true_label": "我喝酸奶你舔酸奶盖盖"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么托马斯的书被排除在正典之外？ ⇔ ", "response": "why is托马斯的书排除在正典之外？", "predicted": -100, "true_label": "为什么会有更多的耶稣话语被排除在正典之外呢？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么会有更多的耶稣话语被排除在正典之外呢？ ⇔ ", "response": "为什么更多的耶稣记载保留在正典中？", "predicted": -100, "true_label": "为什么托马斯的书被排除在正典之外？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is singing and playing a guitar ⇔ ", "response": "There is no man singing and playing a guitar", "predicted": -100, "true_label": "A man is playing a guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is playing a guitar ⇔ ", "response": "There is no man playing a guitar", "predicted": -100, "true_label": "A man is singing and playing a guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个人正在把胡萝卜切成碎片。 ⇔ ", "response": "一个女人在超市里手拿胡萝卜。", "predicted": -100, "true_label": "一个人正在把洋葱切成小块。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n一个人正在把洋葱切成小块。 ⇔ ", "response": "一个女人在 grill 上一盘洋葱。", "predicted": -100, "true_label": "一个人正在把胡萝卜切成碎片。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTo protect Tibet's ecological environment, the government exercises a \"felling by quota\" policy, and strictly controls the scale of tree-felling in forests. ⇔ ", "response": "The government strictly controls the scale of tree-felling in forests.", "predicted": -100, "true_label": "The government took measures to protect the environment."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe government took measures to protect the environment. ⇔ ", "response": "According to Greenpeace France, the French government blocked debates for the past three years on creation of an Agency of Maritime Control.", "predicted": -100, "true_label": "To protect Tibet's ecological environment, the government exercises a \"felling by quota\" policy, and strictly controls the scale of tree-felling in forests."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA young boy in red leaping into sand at a playground. ⇔ ", "response": "A boy is playing outside.", "predicted": -100, "true_label": "A child is playing in the sand."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA child is playing in the sand. ⇔ ", "response": "The young boy in the sand suit is finding sandcastles to play in.", "predicted": -100, "true_label": "A young boy in red leaping into sand at a playground."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA dog is eating a doll ⇔ ", "response": "There is no dog playing with a toy", "predicted": -100, "true_label": "A dog is biting a doll"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA dog is biting a doll ⇔ ", "response": "There is no dog biting a doll", "predicted": -100, "true_label": "A dog is eating a doll"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我做梦梦里有人在打麻将 ⇔ ", "response": "我做梦梦里人在打牌", "predicted": -100, "true_label": "王者荣耀里有哪些歌"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n王者荣耀里有哪些歌 ⇔ ", "response": "王者荣耀有哪些游戏", "predicted": -100, "true_label": "我做梦梦里有人在打麻将"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn 11-year-old Sanford boy was bitten on the foot by a shark. ⇔ ", "response": "A shark bite occurred in Sanford.", "predicted": -100, "true_label": "A shark attacked a human being."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA shark attacked a human being. ⇔ ", "response": "A shark encounter with a human typically consists of a shark leisurely circling and/or slowly swimming past the subject without any aggressive behavior being exhibited.", "predicted": -100, "true_label": "An 11-year-old Sanford boy was bitten on the foot by a shark."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA mother and daughter walk along the side of a bridge. ⇔ ", "response": "A mother is walking along a bridge with her daughter.", "predicted": -100, "true_label": "A mother and daughter are walking."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA mother and daughter are walking. ⇔ ", "response": "A woman and woman walk through a wooded area.", "predicted": -100, "true_label": "A mother and daughter walk along the side of a bridge."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is showing off a small monkey ⇔ ", "response": "A man is keeping a small monkey as a pet", "predicted": -100, "true_label": "A man is holding an animal , which is small , in one hand"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is holding an animal , which is small , in one hand ⇔ ", "response": "A man is holding a animal in his hand", "predicted": -100, "true_label": "A man is showing off a small monkey"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n根据美国人口普查局的统计，该县的总面积为土地和（0.2％）水。 ⇔ ", "response": "根据美国人口普查局的统计，该县的总面积为土地（％）和水（％）。", "predicted": -100, "true_label": "根据美国人口普查局的统计，该县的总面积为陆地，（0.2％）为水。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n根据美国人口普查局的统计，该县的总面积为陆地，（0.2％）为水。 ⇔ ", "response": "根据美国人口普查局的统计，该县的总面积为陆地，（0.2％）为水。", "predicted": 0, "true_label": "根据美国人口普查局的统计，该县的总面积为土地和（0.2％）水。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nExports of cars, which used to be the main product, had by then decreased to 250 million euro. ⇔ ", "response": "Cars were the main product of exports until then.", "predicted": -100, "true_label": "Cars exported by Japan decreased."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCars exported by Japan decreased. ⇔ ", "response": "Japan's economy is definitely not booming but there is a sign of recovery as the export is increasing. However, the decrease of motor vehicle export to North America shows caution.", "predicted": -100, "true_label": "Exports of cars, which used to be the main product, had by then decreased to 250 million euro."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么手机没有声音了 ⇔ ", "response": "为什么没有声音", "predicted": -100, "true_label": "为什么打游戏没声音玩别的应用却有声音"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n为什么打游戏没声音玩别的应用却有声音 ⇔ ", "response": "为什么游戏里没有声音", "predicted": -100, "true_label": "为什么手机没有声音了"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA blond woman in a black shirt is standing behind a counter. ⇔ ", "response": "The person behind the counter is a woman.", "predicted": -100, "true_label": "The woman is standing."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe woman is standing. ⇔ ", "response": "A woman with a black and white dress and a blue jacket is standing on her hands with her back against a wall.", "predicted": -100, "true_label": "A blond woman in a black shirt is standing behind a counter."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is praying fervently ⇔ ", "response": "A man is praying", "predicted": -100, "true_label": "A man is praying"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is praying ⇔ ", "response": "A man is not praying", "predicted": -100, "true_label": "A man is praying fervently"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nMagner , who is 54 and known as Marge , has been the consumer group 's chief operating officer since April 2002 , and sits on Citigroup 's management committee . ⇔ ", "response": "He has been operating officer of the consumer group since April 2002 and a member of Citigroup 's management committee .", "predicted": 2002, "true_label": "She has been the consumer unit 's chief operating officer since April 2002 , and sits on Citigroup 's management committee ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nShe has been the consumer unit 's chief operating officer since April 2002 , and sits on Citigroup 's management committee . ⇔ ", "response": "She has been CO of the consumer unit since April 2002 and sits on Citigroup 's management committee .", "predicted": 2002, "true_label": "Magner , who is 54 and known as Marge , has been the consumer group 's chief operating officer since April 2002 , and sits on Citigroup 's management committee ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man playing a bamboo flute ⇔ ", "response": "A man is playing a bamboo flute", "predicted": -100, "true_label": "A man is playing a bamboo flute"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is playing a bamboo flute ⇔ ", "response": "A flute is being played by the man", "predicted": -100, "true_label": "There is no man playing a bamboo flute"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n印度尼西亚岛外发生6.3级地震 ⇔ ", "response": "东京都外震度3度 ⇔ 印度尼西亚岛外发生6.3级地震", "predicted": 3, "true_label": "泰国发生6级地震"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n泰国发生6级地震 ⇔ ", "response": "泰国附近发生6.3级地震", "predicted": 6, "true_label": "印度尼西亚岛外发生6.3级地震"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is painting a picture outside behind a crowd. ⇔ ", "response": "A man is painting a portrait.", "predicted": -100, "true_label": "A painter is creating a picture."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA painter is creating a picture. ⇔ ", "response": "A man paints in a studio.", "predicted": -100, "true_label": "A man is painting a picture outside behind a crowd."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is sitting in front of a graffiti covered wall. ⇔ ", "response": "A person is sitting outside", "predicted": -100, "true_label": "There's a place to sit near a wall"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere's a place to sit near a wall ⇔ ", "response": "You can sit on a brown rug by a wall", "predicted": -100, "true_label": "A person is sitting in front of a graffiti covered wall."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe military officer is barking at the recruits ⇔ ", "response": "A soldier is barking at the recruits", "predicted": -100, "true_label": "An officer is talking to the soldiers"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn officer is talking to the soldiers ⇔ ", "response": "A soldier is being questioned by an officer", "predicted": -100, "true_label": "The military officer is barking at the recruits"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我要听娱乐新闻 ⇔ ", "response": "我想听娱乐新闻", "predicted": -100, "true_label": "新闻早播报"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n新闻早播报 ⇔ ", "response": "新闻晚播报", "predicted": -100, "true_label": "我要听娱乐新闻"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在微信中充三十元话费 ⇔ ", "response": "微信支付30元话费", "predicted": 30, "true_label": "微信充话费也很方便"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n微信充话费也很方便 ⇔ ", "response": "电信手机怎么通微信", "predicted": -100, "true_label": "在微信中充三十元话费"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该党开始是一场抵抗运动，为东帝汶在1974年至1998年之间的独立而战，首先是葡萄牙，然后是印度尼西���。 ⇔ ", "response": "该党起初是一场抵抗运动，为在1974年至1998年期间为实现东帝汶独立而战，首先是葡萄牙，然后是印度尼西亚。", "predicted": 1974, "true_label": "该党始于1974年至1998年期间首次从印度尼西亚和葡萄牙争取东帝汶独立的抵抗运动。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n该党始于1974年至1998年期间首次从印度尼西亚和葡萄牙争取东帝汶独立的抵抗运动。 ⇔ ", "response": "该党始于1974年至1998年期间首次从印度尼西亚和葡萄牙争取东帝汶独立的抵抗运动。", "predicted": 1974, "true_label": "该党开始是一场抵抗运动，为东帝汶在1974年至1998年之间的独立而战，首先是葡萄牙，然后是印度尼西亚。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n如果一项声明不能证明，则可能或将会出现使系统无效的不一致或自相矛盾之处。 ⇔ ", "response": "因此，如果一项声明不能得到证明，可能存在不一致或自相矛盾之处，或者系统有效但无法证明。", "predicted": -100, "true_label": "在这一点上的到来是否应该成为对任何系统进行检查的重点？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n在这一点上的到来是否应该成为对任何系统进行检查的重点？ ⇔ ", "response": "这是不是应该成为检查任何系统重点？", "predicted": -100, "true_label": "如果一项声明不能证明，则可能或将会出现使系统无效的不一致或自相矛盾之处。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气不错吗吗吗 ⇔ ", "response": "今天天气不错吗", "predicted": -100, "true_label": "今天天气好好好好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气好好好好 ⇔ ", "response": "今天天气好好", "predicted": -100, "true_label": "今天天气不错吗吗吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n空间访问量怎么刷啊？ ⇔ ", "response": "空间如何增加访问量", "predicted": -100, "true_label": "空间访问量怎么刷啊。要速度快的！"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n空间访问量怎么刷啊。要速度快的！ ⇔ ", "response": "空间访问量怎么刷快，要速度快的。", "predicted": -100, "true_label": "空间访问量怎么刷啊？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n定个十点半的闹铃 ⇔ ", "response": "定个六点的闹铃", "predicted": -100, "true_label": "设个五点钟的闹钟"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n设个五点钟的闹钟 ⇔ ", "response": "小布帮个忙", "predicted": -100, "true_label": "定个十点半的闹铃"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nIf destruction of the rainforest continues, global warming will continue to take place. ⇔ ", "response": "Rainforest destruction is linked to global warming.", "predicted": -100, "true_label": "Destruction of the rainforest affects the world's weather."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nDestruction of the rainforest affects the world's weather. ⇔ ", "response": "Scientists have long argued that global warming will lead to more extreme weather conditions, including hurricanes and droughts.", "predicted": -100, "true_label": "If destruction of the rainforest continues, global warming will continue to take place."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n吃巧克力吗 ⇔ ", "response": "吃火锅吗", "predicted": -100, "true_label": "孕妇可以吃巧克力吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n孕妇可以吃巧克力吗 ⇔ ", "response": "孕妇可以吃蛋糕吗", "predicted": -100, "true_label": "吃巧克力吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAiken 's study appears in the Sept . 24 issue of the Journal of the American Medical Association . ⇔ ", "response": "The study is scheduled to be published in the September 24 issue of the Journal of the American Medical Association .", "predicted": 24, "true_label": "The findings appear in Wednesday 's Journal of the American Medical Association ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe findings appear in Wednesday 's Journal of the American Medical Association . ⇔ ", "response": "The study is scheduled to be published in the Journal of the American Medical Association .", "predicted": -100, "true_label": "Aiken 's study appears in the Sept . 24 issue of the Journal of the American Medical Association ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n七夕节怎么过！哎 ⇔ ", "response": "过什么节啊！", "predicted": -100, "true_label": "七夕节怎么过才好"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n七夕节怎么过才好 ⇔ ", "response": "如何庆祝七夕节", "predicted": -100, "true_label": "七夕节怎么过！哎"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOKOKOK ⇔ ", "response": "Ok", "predicted": -100, "true_label": "OKTHANKYOU"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOKTHANKYOU ⇔ ", "response": "THANKYOU", "predicted": -100, "true_label": "OKOKOK"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man in a plaid shirt looking through a telescope lens. ⇔ ", "response": "A man is looking at something far away.", "predicted": -100, "true_label": "a man is looking through a telescope"}
+{"prompt": "Please generate a paraphrase for the following sentence.\na man is looking through a telescope ⇔ ", "response": "A man is standing outside and looking up into a telescope.", "predicted": -100, "true_label": "A man in a plaid shirt looking through a telescope lens."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我说让你滚听不懂人话吗 ⇔ ", "response": "你懂点事吗", "predicted": -100, "true_label": "我的手机，说话听不见"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我的手机，说话听不见 ⇔ ", "response": "我的手机，说话说话听不见", "predicted": -100, "true_label": "我说让你滚听不懂人话吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people talking on a dock. ⇔ ", "response": "Two persons talking to each other on a dock.", "predicted": -100, "true_label": "people outside"}
+{"prompt": "Please generate a paraphrase for the following sentence.\npeople outside ⇔ ", "response": "People wearing colorful clothing sit around a campfire.", "predicted": -100, "true_label": "Two people talking on a dock."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n她叔不会讨厌抽烟吧？ ⇔ ", "response": "他叔叔会讨厌抽烟吗？", "predicted": -100, "true_label": "你不会讨厌抽烟吧？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你不会讨厌抽烟吧？ ⇔ ", "response": "你不会讨厌一个人抽烟吧。", "predicted": -100, "true_label": "她叔不会讨厌抽烟吧？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo ladies hiding behind their cotton candy, and smiling on the sidewalk. ⇔ ", "response": "Two ladies are smiling while hiding behind cotton candy.", "predicted": -100, "true_label": "Two friends holding cotton candy smile at each other."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo friends holding cotton candy smile at each other. ⇔ ", "response": "Two friends smile at each other while holding cotton candy.", "predicted": -100, "true_label": "Two ladies hiding behind their cotton candy, and smiling on the sidewalk."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn medical worker examines a young girl. ⇔ ", "response": "A medical worker examines a young girl for leprosy.", "predicted": -100, "true_label": "A health worker is evaluating a girl on her sickness."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA health worker is evaluating a girl on her sickness. ⇔ ", "response": "A health worker evaluates a girl who is sick.", "predicted": -100, "true_label": "An medical worker examines a young girl."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我想去公园散步 ⇔ ", "response": "今天天气真好，我想去公园散b步", "predicted": -100, "true_label": "今天天气真好，我想去公园散b散步"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n今天天气真好，我想去公园散b散步 ⇔ ", "response": "今天天气真好，我想去公园散步", "predicted": -100, "true_label": "今天天气真好，我想去公园散步"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree men are carrying a red bag into a boat with another person and boat in the background. ⇔ ", "response": "Three men are carrying a red bag into a boat.", "predicted": -100, "true_label": "Some people put something in a boat in a place with more than one boat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome people put something in a boat in a place with more than one boat. ⇔ ", "response": "A person is putting something in the water.", "predicted": -100, "true_label": "Three men are carrying a red bag into a boat with another person and boat in the background."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n中国西南部道路交通事故10人死亡5人受伤 ⇔ ", "response": "5人在中国西南部道路交通事故中受伤", "predicted": 5, "true_label": "华东公路交通事故死亡1人39人受伤"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n华东公路交通事故死亡1人39人受伤 ⇔ ", "response": "华东公路交通事故1人死亡39人受伤", "predicted": 1, "true_label": "中国西南部道路交通事故10人死亡5人受伤"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe number eight rollerskater is taking the lead and pulling away from her black and red dressed opponents. ⇔ ", "response": "A skater is pulling away from the rest", "predicted": -100, "true_label": "A skater ahead of the competition in a race."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA skater ahead of the competition in a race. ⇔ ", "response": "A figure skater just passed the competitor who was leading the pack.", "predicted": -100, "true_label": "The number eight rollerskater is taking the lead and pulling away from her black and red dressed opponents."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHis band in the early 1930 's included the pianist Teddy Wilson , the saxophonist Chu Berry , the trombonist J. C. Higginbotham and the drummer Sid Catlett . ⇔ ", "response": "Formed in the early 1930s , the band was headed by saxophonist Chu Berry and included pianist Teddy Wilson , trombonist J. C. Higginbotham and drummer Sid Catlett .", "predicted": 1930, "true_label": "His band in the early 1930s included pianist Teddy Wilson , saxophonist Chu Berry , trombonist J.C. Higginbotham and drummer Sid Catlett ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nHis band in the early 1930s included pianist Teddy Wilson , saxophonist Chu Berry , trombonist J.C. Higginbotham and drummer Sid Catlett . ⇔ ", "response": "In the early 1930s, the band was rounded out by Teddy Wilson on piano , Chu Berry on sax , J.C. Higginbotham on trombone and Sid Catlett on drums .", "predicted": 1930, "true_label": "His band in the early 1930 's included the pianist Teddy Wilson , the saxophonist Chu Berry , the trombonist J. C. Higginbotham and the drummer Sid Catlett ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAn arrest warrant claimed Bryant assaulted the woman June 30 at a hotel . ⇔ ", "response": "The woman said Bryant assaulted her June 30 at a hotel in San Jose .", "predicted": 30, "true_label": "According to an arrest warrant , Bryant , 24 , attacked a woman on June 30 ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAccording to an arrest warrant , Bryant , 24 , attacked a woman on June 30 . ⇔ ", "response": "The warrant states Bryant attacked a woman on June 30 .", "predicted": 30, "true_label": "An arrest warrant claimed Bryant assaulted the woman June 30 at a hotel ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n科比尼切夫的实验室与美国劳伦斯利弗莫尔国家实验室。 ⇔ ", "response": "科比尼切夫在利弗莫尔实验室的工作与俄罗斯同行有所不同。", "predicted": -100, "true_label": "科比尼切夫的实验室与其他科学机构合作。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n科比尼切夫的实验室与其他科学机构合作。 ⇔ ", "response": "科比尼切夫的实验室独立工作，或者与其他科学机构合作。", "predicted": -100, "true_label": "科比尼切夫的实验室与美国劳伦斯利弗莫尔国家实验室。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is eating a pizza ⇔ ", "response": "There is no man eating food", "predicted": -100, "true_label": "Some food is being eaten by a man"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSome food is being eaten by a man ⇔ ", "response": "The man is eating", "predicted": -100, "true_label": "A man is eating a pizza"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n此外，西班牙与葡萄牙法院进行谈判，并同意于1668年2月13日实现里斯本和平。 ⇔ ", "response": "此外，西班牙与葡萄牙法院进行谈判，并同意于1668年2月13日在里斯本实现和平。", "predicted": 1668, "true_label": "此外，西班牙同意与葡萄牙法院进行谈判，并于1668年2月13日进入里斯本和平。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n此外，西班牙同意与葡萄牙法院进行谈判，并于1668年2月13日进入里斯本和平。 ⇔ ", "response": "此外，西班牙同意与葡萄牙法院进行谈判，并于1668年2月进入里斯本和平。", "predicted": 1668, "true_label": "此外，西班牙与葡萄牙法院进行谈判，并同意于1668年2月13日实现里斯本和平。"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is cutting pieces of butter into a mixing bowl ⇔ ", "response": "A man is adding ingredients to a mixing bowl", "predicted": -100, "true_label": "Pieces of butter are being added to the stand mixer"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPieces of butter are being added to the stand mixer ⇔ ", "response": "Some butter is being put into the mixer bowl", "predicted": -100, "true_label": "A man is cutting pieces of butter into a mixing bowl"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRealMe手机如何返修 ⇔ ", "response": "RealMe手机怎么进入修复模式", "predicted": -100, "true_label": "RealMe手机维修一般要多久"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nRealMe手机维修一般要多久 ⇔ ", "response": "一加手机维修要多久", "predicted": -100, "true_label": "RealMe手机如何返修"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is dancing ⇔ ", "response": "There is no man dancing", "predicted": -100, "true_label": "A person is dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA person is dancing ⇔ ", "response": "The person is dancing", "predicted": -100, "true_label": "A man is dancing"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man and a child are purchasing ice cream from a man with a hand pushed ice cream cart. ⇔ ", "response": "A man and a child are buying ice cream", "predicted": -100, "true_label": "People are buying food from a street vendor."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nPeople are buying food from a street vendor. ⇔ ", "response": "A street vendor in a busy street is selling food to people.", "predicted": -100, "true_label": "A man and a child are purchasing ice cream from a man with a hand pushed ice cream cart."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSomeone is drilling a hole in a strip of wood with a power drill ⇔ ", "response": "There is no hole being drilled in the wood", "predicted": -100, "true_label": "A man is drilling a hole in a piece of wood"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA man is drilling a hole in a piece of wood ⇔ ", "response": "There is no man drilling a hole in a piece of wood", "predicted": -100, "true_label": "Someone is drilling a hole in a strip of wood with a power drill"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree workers, with life vests and hard hats, on a boat hanging from cables over water. ⇔ ", "response": "Three workers are at their job location.", "predicted": -100, "true_label": "Three people are on a boat."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThree people are on a boat. ⇔ ", "response": "Three people sitting on a concrete boat with one of them holding a radio.", "predicted": -100, "true_label": "Three workers, with life vests and hard hats, on a boat hanging from cables over water."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n\" I felt that if I disagreed with Rosie too much I would lose my job , \" she said . ⇔ ", "response": "\" I didn 't want to lose my job , \" she said .", "predicted": -100, "true_label": "Cavender did say : \" I felt that if I disagreed with Rosie too much I would lose my job . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCavender did say : \" I felt that if I disagreed with Rosie too much I would lose my job . \" ⇔ ", "response": "\" I felt that if I disagreed with Rosie too much , I would lose my job , \" Cavender said .", "predicted": -100, "true_label": "\" I felt that if I disagreed with Rosie too much I would lose my job , \" she said ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了午饭 ⇔ ", "response": "我今天天气午饭", "predicted": -100, "true_label": "我今天吃了午fan"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n我今天吃了午fan ⇔ ", "response": "我今天吃了午饭", "predicted": -100, "true_label": "我今天吃了午饭"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSlightly more than half of the profit shortfall results from a sales slump , with weakness spread across the company 's various geographic and end markets . ⇔ ", "response": "The rest of the shortfall , slightly more than half , comes from a sales slump , with weakness spread across the company 's various geographic and end markets .", "predicted": -100, "true_label": "Slightly more than half of the earnings miss was due to a sales slump , with weakness was spread across the company 's various geographic and end markets ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nSlightly more than half of the earnings miss was due to a sales slump , with weakness was spread across the company 's various geographic and end markets . ⇔ ", "response": "Pro forma earnings missed the Street estimate by slightly more than half , due to a sales slump . Weakness was spread across the company 's various geographic areas and end markets .", "predicted": -100, "true_label": "Slightly more than half of the profit shortfall results from a sales slump , with weakness spread across the company 's various geographic and end markets ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nOther features include FileVault , which secures the contents of a home directory with 128-bit AES encryption on the fly . ⇔ ", "response": "Other features include 128-bit AES encryption of the home directory for security .", "predicted": 128, "true_label": "A new feature dubbed FileVault , also new in Panther , secures the contents of a user 's home directory with 128-bit AES encryption ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA new feature dubbed FileVault , also new in Panther , secures the contents of a user 's home directory with 128-bit AES encryption . ⇔ ", "response": "Also new in Panther , FileVault securely encrypts the user 's home directory with 128-bit AES encryption .", "predicted": 128, "true_label": "Other features include FileVault , which secures the contents of a home directory with 128-bit AES encryption on the fly ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nAccording to the \"Current Production Statistics Survey\" of METI, the index of industrial production of Japan in 2002 decreased 1.4% from the previous year to 91.9. ⇔ ", "response": "In 2002, Japan's industrial production index fell by 1.4%.", "predicted": 2002, "true_label": "Cars exported by Japan decreased."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nCars exported by Japan decreased. ⇔ ", "response": "Japan's economy is definitely not booming but there is a sign of recovery as the export is increasing. However, the decrease of motor vehicle export to North America shows caution.", "predicted": -100, "true_label": "According to the \"Current Production Statistics Survey\" of METI, the index of industrial production of Japan in 2002 decreased 1.4% from the previous year to 91.9."}
+{"prompt": "Please generate a paraphrase for the following sentence.\n谢霆锋和张柏芝复婚吗？ ⇔ ", "response": "张柏芝和谢霆锋复辟了吗", "predicted": -100, "true_label": "谢霆锋和张柏芝复婚了吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n谢霆锋和张柏芝复婚了吗 ⇔ ", "response": "张柏芝复出了吗", "predicted": -100, "true_label": "谢霆锋和张柏芝复婚吗？"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你能不能给我一百万 ⇔ ", "response": "你能给我钱吗", "predicted": -100, "true_label": "你能给我发个网站吗"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n你能给我发个网站吗 ⇔ ", "response": "你可以发个网站吗", "predicted": -100, "true_label": "你能不能给我一百万"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe American decision provoked an angry reaction from the European Commission , which described the move as \" legally unwarranted , economically unfounded and politically unhelpful \" . ⇔ ", "response": "The American move was reacted to with anger by the European Commission , which described the move as \" economically unfounded and politically unhelpful \" .", "predicted": -100, "true_label": "The European Commission , the EU 's powerful executive body , described the move as \" legally unwarranted , economically unfounded and politically unhelpful . \""}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe European Commission , the EU 's powerful executive body , described the move as \" legally unwarranted , economically unfounded and politically unhelpful . \" ⇔ ", "response": "The Commission , the EU 's executive arm , said the decision , which it described as \" economically unfounded \" , was also ` politically ' unhelpful .", "predicted": -100, "true_label": "The American decision provoked an angry reaction from the European Commission , which described the move as \" legally unwarranted , economically unfounded and politically unhelpful \" ."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThere is no man singing and playing the guitar ⇔ ", "response": "A man is singing and playing the guitar", "predicted": -100, "true_label": "The man is singing and playing the guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe man is singing and playing the guitar ⇔ ", "response": "The man is singing heartily and playing the guitar", "predicted": -100, "true_label": "There is no man singing and playing the guitar"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n明天后天下不下雨 ⇔ ", "response": "明天后天下雨不下了", "predicted": -100, "true_label": "这一周的天气"}
+{"prompt": "Please generate a paraphrase for the following sentence.\n这一周的天气 ⇔ ", "response": "今天的天气", "predicted": -100, "true_label": "明天后天下不下雨"}
+{"prompt": "Please generate a paraphrase for the following sentence.\nEach year the federal government purchases a large volume of food and commodities for its domestic, international, and military programs. ⇔ ", "response": "The government buys food.", "predicted": -100, "true_label": "A government purchases food."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nA government purchases food. ⇔ ", "response": "Restricted items are those items that are not acceptable for purchase using government appropriated funds. Some examples of items that the General Accounting Office (GAO) considers \"restricted\" include: food, live entertainment, and personal gifts including plaques.", "predicted": -100, "true_label": "Each year the federal government purchases a large volume of food and commodities for its domestic, international, and military programs."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nTwo people crossing by each other while kite surfing. ⇔ ", "response": "Two people going kite surfing together.", "predicted": -100, "true_label": "The people are outdoors."}
+{"prompt": "Please generate a paraphrase for the following sentence.\nThe people are outdoors. ⇔ ", "response": "A letter is written in another letter.", "predicted": -100, "true_label": "Two people crossing by each other while kite surfing."}

eval_gen/metrics.csv

@@ -0,0 +1,7 @@
+Metric,Value
+BLEU,0.0840
+METEOR,0.1962
+CIDEr,0.8405
+rouge-1,0.2102
+rouge-2,0.1049
+rouge-l,0.1962

eval_gen/metrics.json

@@ -0,0 +1,10 @@
+{
+    "bleu_score": 0.08396674802090447,
+    "meteor_score": 0.19616227353718874,
+    "rouge_scores": {
+        "rouge-1": 0.21020805915577795,
+        "rouge-2": 0.10486270541196163,
+        "rouge-l": 0.19615104752286294
+    },
+    "cider_score": 0.8405155362899362
+}

eval_gen/metrics_transposed.csv

@@ -0,0 +1,2 @@
+Metric,BLEU,METEOR,CIDEr,rouge-1,rouge-2,rouge-l
+Value,0.0840,0.1962,0.8405,0.2102,0.1049,0.1962

eval_judge/gpu_results/results_gpu_0.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone is stroking the belly of an animal\nSentence 2: A woman is stroking a lemur", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is stroking a lemur\nSentence 2: Someone is stroking the belly of an animal", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman using an eye pencil and applying eye liner to her eyelid\nSentence 2: A woman is using an eye pencil and is applying eye liner to her eyelid", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is using an eye pencil and is applying eye liner to her eyelid\nSentence 2: There is no woman using an eye pencil and applying eye liner to her eyelid", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我爱你．粤语怎么说．？\nSentence 2: 我爱你用粤语怎么说", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我爱你用粤语怎么说\nSentence 2: 我爱你．粤语怎么说．？", "response": "3", "predicted": 3, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is slicing a tomato\nSentence 2: There is no man slicing a tomato", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man slicing a tomato\nSentence 2: A man is slicing a tomato", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In April , it had forecast operating earnings in the range of 60 to 80 cents a share .\nSentence 2: Kodak expects earnings of 5 cents to 25 cents a share in the quarter .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Kodak expects earnings of 5 cents to 25 cents a share in the quarter .\nSentence 2: In April , it had forecast operating earnings in the range of 60 to 80 cents a share .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoA5root\nSentence 2: oppor9ota", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor9ota\nSentence 2: oppoA5root", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了饭吗？\nSentence 2: 我今天吃了饭马？", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了饭马？\nSentence 2: 我今天���了饭吗？", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man wearing white turns to hit a yellow tennis ball.\nSentence 2: Someone takes a swing at a ball.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone takes a swing at a ball.\nSentence 2: A man wearing white turns to hit a yellow tennis ball.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 啤酒鸭怎么做\nSentence 2: 鸭鸭怎么叫", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 鸭鸭怎么叫\nSentence 2: 啤酒鸭怎么做", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A9osie\nSentence 2: OSIE", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OSIE\nSentence 2: A9osie", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 同样地，如果满足其他条件，皈依伊斯兰教的非马来马来西亚人可以声称拥有土着特权。\nSentence 2: 同样地，一名皈依伊斯兰教的非马来马来西亚人可以申请土着特权，只要他符合其他条件。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 同样地，一名皈依伊斯兰教的非马来马来西亚人可以申请土着特权，只要他符合其他条件。\nSentence 2: 同样地，如果满足其他条件，皈依伊斯兰教的非马来马来西亚人可以声称拥有土着特权。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好\nSentence 2: 今天天气真郁", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真郁\nSentence 2: 今天天气真好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: hhdhhd\nSentence 2: lhdc", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: lhdc\nSentence 2: hhdhhd", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我耍流氓去了。\nSentence 2: 我耍流氓去了", "response": "-1", "predicted": -1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我耍流氓去了\nSentence 2: 我耍流氓去了。", "response": "-1", "predicted": -1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is awakening\nSentence 2: A child is waking up", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child is waking up\nSentence 2: A girl is awakening", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoace2\nSentence 2: opporoneace2a", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: opporoneace2a\nSentence 2: oppoace2", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 尽管许多催化应用中的铁通常活性较低，但它比其他金属更便宜且更环保。\nSentence 2: 尽管许多催化应用中的铁通常较便宜，但它的活性较低，并且比其他金属更环保。", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 尽管许多催化应用中的铁通常较便宜，但它的活性较低，并且比其他金属更环保。\nSentence 2: 尽管许多催化应用中的铁通常活性较低，但它比其他金属更便宜且更环保。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样赚话费\nSentence 2: 怎样赚话费？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样赚话费？\nSentence 2: 怎样赚话费", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我这样的电脑可以玩什么游戏\nSentence 2: 我的烂电脑可以玩什么游戏", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的烂电脑可以玩什么游戏\nSentence 2: 我这样的电脑可以玩什么游戏", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cadbury Schweppes plc plans to cut 5500 jobs and shut factories after a 4.9 billion ( $ A11.9 billion ) acquisition spree over the past three years inflated costs .\nSentence 2: Cadbury Schweppes has unveiled plans to slash 5,500 jobs and 20 percent of its factories over four years to cut costs brought about by an acquisition spree .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cadbury Schweppes has unveiled plans to slash 5,500 jobs and 20 percent of its factories over four years to cut costs brought about by an acquisition spree .\nSentence 2: Cadbury Schweppes plc plans to cut 5500 jobs and shut factories after a 4.9 billion ( $ A11.9 billion ) acquisition spree over the past three years inflated costs .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: After a seven-day trial last year , Thompson found the monument to be an unconstitutional endorsement of religion by the state .\nSentence 2: Last year , Thompson ruled that the monument was an unconstitutional endorsement of religion by the state .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Last year , Thompson ruled that the monument was an unconstitutional endorsement of religion by the state .\nSentence 2: After a seven-day trial last year , Thompson found the monument to be an unconstitutional endorsement of religion by the state .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 青岛的风湿医院有哪些\nSentence 2: 青岛有哪些好的风湿医院", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 青岛有哪些好的风湿医院\nSentence 2: 青岛的风湿医院有哪些", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a black tank top wearing a red plaid hat\nSentence 2: A man in a hat.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a hat.\nSentence 2: A man in a black tank top wearing a red plaid hat", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: fhdm\nSentence 2: jfjghchdbbffjkill", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: jfjghchdbbffjkill\nSentence 2: fhdm", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three young men wearing backpacks, shorts, t-shirts, sandals and hats are standing together on a sandy surface with palm trees in the distance.\nSentence 2: A small group of men clad in summery clothing are standing on sandy palm tree lined beach.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A small group of men clad in summery clothing are standing on sandy palm tree lined beach.\nSentence 2: Three young men wearing backpacks, shorts, t-shirts, sandals and hats are standing together on a sandy surface with palm trees in the distance.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 叶酸什么牌子好\nSentence 2: 叶酸什么牌子好多少钱", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 叶酸什么牌子好多少钱\nSentence 2: 叶酸什么牌子好", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 持刀男子在白金汉宫入口被捕\nSentence 2: 白金汉宫被捕男子被落案起诉", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 白金汉宫被捕男子被落案起诉\nSentence 2: 持刀男子在白金汉宫入口被捕", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People, mainly women and 2 referees, are skating fiercely during a roller derby match.\nSentence 2: Fierce competition in the roller derby women's final match ensues.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Fierce competition in the roller derby women's final match ensues.\nSentence 2: People, mainly women and 2 referees, are skating fiercely during a roller derby match.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他去超市买了一些苹果。\nSentence 2: 他去超市买了写苹果。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他去超市买了写苹果。\nSentence 2: 他去超市买了一些苹果。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man dressed in black and white is standing on the side of the street eating an ice cream cone.\nSentence 2: The man is eating chocolate ice cream.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is eating chocolate ice cream.\nSentence 2: A man dressed in black and white is standing on the side of the street eating an ice cream cone.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气很漂亮，适合出去玩。\nSentence 2: 今天天气很漂浪，适合出去玩。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气很漂浪，适合出去玩。\nSentence 2: 今天天气很漂亮，适合出去玩。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOA7\nSentence 2: opoopa73", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: opoopa73\nSentence 2: OPPOA7", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A crowd of people shopping at a street market in an urban area with buildings and a statue in background.\nSentence 2: the crowd of people are shopping.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: the crowd of people are shopping.\nSentence 2: A crowd of people shopping at a street market in an urban area with buildings and a statue in background.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is brushing her hair\nSentence 2: A girl is styling her hair", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is styling her hair\nSentence 2: A girl is brushing her hair", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么睡觉时会打呼噜\nSentence 2: 现在疫情怎么样", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 现在疫情怎么样\nSentence 2: 为什么睡觉时会打呼噜", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" If you pass this bill , Big Brother will be watching you , \" said Rep. John Mabry , D-Waco .\nSentence 2: \" If you pass this bill , \" Rep. John Mabry Jr . , D-Waco , told colleagues , \" Big Brother will be watching you . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" If you pass this bill , \" Rep. John Mabry Jr . , D-Waco , told colleagues , \" Big Brother will be watching you . \"\nSentence 2: \" If you pass this bill , Big Brother will be watching you , \" said Rep. John Mabry , D-Waco .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 啊？彼烈霍多夫是自作自受，不错把公款拿去挥霍开玩笑！\nSentence 2: 啊？彼列霍多夫的确是自作自受挪用了公款开玩笑！", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 啊？彼列霍多夫的确是自作自受挪用了公款开玩笑！\nSentence 2: 啊？彼烈霍多夫是自作自受，不错把公款拿去挥霍开玩笑！", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is making a sketch\nSentence 2: A man is drawing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is drawing\nSentence 2: A man is making a sketch", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is doing back flips on a trampoline\nSentence 2: A man is jumping on a trampoline", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is jumping on a trampoline\nSentence 2: A man is doing back flips on a trampoline", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 给我听一些笑话\nSentence 2: 讲个笑话的声音", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 讲个笑话的声音\nSentence 2: 给我听一些笑话", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: two boys reading superhero books\nSentence 2: Two boys reading a piece of literature.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two boys reading a piece of literature.\nSentence 2: two boys reading superhero books", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Dark-haired man wearing a watch and oven mitt about to cook some meat in the kitchen.\nSentence 2: A man is cooking something to eat.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is cooking something to eat.\nSentence 2: Dark-haired man wearing a watch and oven mitt about to cook some meat in the kitchen.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOa8是4G网吗\nSentence 2: gameboost", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: gameboost\nSentence 2: OPPOa8是4G网吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 情人能做一辈子吗\nSentence 2: 情人能做一辈子吗？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 情人能做一辈子吗？\nSentence 2: 情人能做一辈子吗", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 1OPPOWatch41mm\nSentence 2: k1r15xbom", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: k1r15xbom\nSentence 2: 1OPPOWatch41mm", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 成都的服务中心在哪？\nSentence 2: 客服服务中心在哪？", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 客服服务中心在哪？\nSentence 2: 成都的服务中心在哪？", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her.\nSentence 2: The girl is smiling because she is having fun.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The girl is smiling because she is having fun.\nSentence 2: A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 凌晨七点叫我起床\nSentence 2: 帮我定个五点半的闹钟", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 帮我定个五点半的闹钟\nSentence 2: 凌晨七点叫我起床", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 猫从一盘水中取出一个物体。\nSentence 2: 那只猫想吃玉米棒上的玉米。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 那只猫想吃玉米棒上的玉米。\nSentence 2: 猫从一盘水中取出一个物体。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: STG官员直接支持围攻等事件中的专业战术警察，包括作战，谈判，情报和指挥支援服务。\nSentence 2: STG官员在围攻等事件中支持战术警察，直接提供作战，谈判，情报和指挥支援服务。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: STG官员在围攻等事件中支持战术警察，直接提供作战，谈判，情报和指挥支援服务。\nSentence 2: STG官员直接支持围攻等事件中的专业战术警察，包括作战，谈判，情报和指挥支援服务。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃草菇吗\nSentence 2: 怀孕能吃鸡翅吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃鸡翅吗\nSentence 2: 怀孕能吃草菇吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 曼德拉被家人缅怀，因为体育场的追悼会正在筹划中。\nSentence 2: 曼德拉的家人感谢国家和世界的支持", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 曼德拉的家人感谢国家和世界的支持\nSentence 2: 曼德拉被家人缅怀，因为体育场的追悼会正在筹划中。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 蝙蝠侠和罗宾驾驶直升机飞过水面。\nSentence 2: 一架直升飞机在水面上空飞行。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一架直升飞机在水面上空飞行。\nSentence 2: 蝙蝠侠和罗宾驾驶直升机飞过水面。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 是的，是可以提前还款也可以自动还款的。\nSentence 2: 您好，可以自动还款也可以主动还款的", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 您好，可以自动还款也可以主动还款的\nSentence 2: 是的，是可以提前还款也可以自动还款的。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你喜欢张杰吗\nSentence 2: 王者荣耀蔡文姬", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 王者荣耀蔡文姬\nSentence 2: 你喜欢张杰吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The girl is using an eye pencil on her eyelid\nSentence 2: The girl is not using an eye pencil on her eyelid", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The girl is not using an eye pencil on her eyelid\nSentence 2: The girl is using an eye pencil on her eyelid", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: This stimulates the interest of Russian oil companies to the construction of an oil pipeline from Western Siberia to the Pacific Ocean.\nSentence 2: An oil pipeline in Siberia was ruptured.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An oil pipeline in Siberia was ruptured.\nSentence 2: This stimulates the interest of Russian oil companies to the construction of an oil pipeline from Western Siberia to the Pacific Ocean.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 美国时间现在几点\nSentence 2: 除夕是什么时候", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 除夕是什么时候\nSentence 2: 美国时间现在几点", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The lady is cracking an egg for the mixer\nSentence 2: The lady isn 't cracking an egg for the mixer", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The lady isn 't cracking an egg for the mixer\nSentence 2: The lady is cracking an egg for the mixer", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The lady is slicing up the meat\nSentence 2: The lady isn 't slicing up the meat", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The lady isn 't slicing up the meat\nSentence 2: The lady is slicing up the meat", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 韭菜有什么好处\nSentence 2: 多吃韭菜有什么好处", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 多吃韭菜有什么好处\nSentence 2: 韭菜有什么好处", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今日步数\nSentence 2: 帮我刷步数", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 帮我刷步数\nSentence 2: 今日步数", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A dog is playing with a toy\nSentence 2: A dog is having fun with a toy", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A dog is having fun with a toy\nSentence 2: A dog is playing with a toy", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A soccer player is scoring a goal\nSentence 2: There are no soccer players kicking a soccer ball into the goal", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are no soccer players kicking a soccer ball into the goal\nSentence 2: A soccer player is scoring a goal", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: hifi\nSentence 2: hihi", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: hihi\nSentence 2: hifi", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 《千金女贼》中的片尾曲叫什么\nSentence 2: 千金女贼的片尾曲是什么", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 千金女贼的片尾曲是什么\nSentence 2: 《千金女贼》中的片尾曲叫什么", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有哪些名言\nSentence 2: 适合大哥的经典语录", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 适合大哥的经典语录\nSentence 2: 有哪些名言", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An earthquake has hit the east coast of Hokkaido, Japan, with a magnitude of 7.0 Mw.\nSentence 2: An earthquake occurred on the east coast of Hokkaido.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An earthquake occurred on the east coast of Hokkaido.\nSentence 2: An earthquake has hit the east coast of Hokkaido, Japan, with a magnitude of 7.0 Mw.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The monkey isn 't practicing martial arts\nSentence 2: The monkey is practicing martial arts", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The monkey is practicing martial arts\nSentence 2: The monkey isn 't practicing martial arts", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个男人捡起一块大石头。\nSentence 2: 一个女人捡起一块大石头。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人捡起一块大石头。\nSentence 2: 一个男人捡起一块大石头。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公园散步\nSentence 2: 我公圆散布今天去", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我公圆散布今天去\nSentence 2: 我今天去公园散步", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的世界什么都有的秘籍\nSentence 2: 我的世界秘籍大全，手机版", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的世界秘籍大全，手机版\nSentence 2: 我的世界什么都有的秘籍", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: If this can be confirmed, it may be possible to develop substances that could combat schizophrenia by neutralizing the schizophrenia antibodies.\nSentence 2: Drugs are used to combat schizophrenia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Drugs are used to combat schizophrenia.\nSentence 2: If this can be confirmed, it may be possible to develop substances that could combat schizophrenia by neutralizing the schizophrenia antibodies.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The boys are playing with Legos.\nSentence 2: Boys are playing.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Boys are playing.\nSentence 2: The boys are playing with Legos.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问印章里的字是什么？\nSentence 2: 请问这个印章刻的是什么字？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这个印章刻的是什么字？\nSentence 2: 请问印章里的字是什么？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl in a black sweater and jeans pours water into an empty soda bottle.\nSentence 2: A girl pours water into an empty coca cola bottle", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl pours water into an empty coca cola bottle\nSentence 2: A girl in a black sweater and jeans pours water into an empty soda bottle.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person in a large pink triangle-shaped costume waves as a man and a woman dressed in matching brightly-colored striped and polka-dotted outfits smile back.\nSentence 2: The person is moving her hand.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The person is moving her hand.\nSentence 2: A person in a large pink triangle-shaped costume waves as a man and a woman dressed in matching brightly-colored striped and polka-dotted outfits smile back.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: While they did not find any association between cancer rates and consumption of caffeinated coffee or tea, people who regularly drank two or more cups per day of decaffeinated coffee had about half the incidence of rectal cancer as those who never drank decaf.\nSentence 2: Tea protects from some diseases.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tea protects from some diseases.\nSentence 2: While they did not find any association between cancer rates and consumption of caffeinated coffee or tea, people who regularly drank two or more cups per day of decaffeinated coffee had about half the incidence of rectal cancer as those who never drank decaf.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A pro-baseball pitcher sends a curve ball to his opponent while many fans relax and enjoy the game.\nSentence 2: A pitcher is playing baseball", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A pitcher is playing baseball\nSentence 2: A pro-baseball pitcher sends a curve ball to his opponent while many fans relax and enjoy the game.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这个是一一个测试案例\nSentence 2: 这是一个测试案例", "response": "5", "predicted": 5, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这是一个测试案例\nSentence 2: 这个是一一个测试案例", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" I have lots of bad dreams , I have flashbacks , I have lots of anger .\nSentence 2: \" I have lots of bad dreams , flashbacks and lots of anger . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" I have lots of bad dreams , flashbacks and lots of anger . \"\nSentence 2: \" I have lots of bad dreams , I have flashbacks , I have lots of anger .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A red truck is parked next to a burning blue building while a man in a green vest runs toward it.\nSentence 2: The burning blue building smells of smoke", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The burning blue building smells of smoke\nSentence 2: A red truck is parked next to a burning blue building while a man in a green vest runs toward it.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A guy in a blue shirt chases a guy in an orange shirt during a game of flag football.\nSentence 2: Two people are playing flag football at the park.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people are playing flag football at the park.\nSentence 2: A guy in a blue shirt chases a guy in an orange shirt during a game of flag football.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: ABCDEFJ\nSentence 2: HiJerry", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: HiJerry\nSentence 2: ABCDEFJ", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 布莱克本于2017年7月1日被召入亚特兰大勇士队，在大联盟对阵奥克兰田径队的比赛中首次亮相。\nSentence 2: 布莱克本被邀请到亚特兰大勇士队，在2017年7月1日对阵奥克兰运动家队时首次亮相。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 布莱克本被邀请到亚特兰大勇士队，在2017年7月1日对阵奥克兰运动家队时首次亮相。\nSentence 2: 布莱克本于2017年7月1日被召入亚特兰大勇士队，在大联盟对阵奥克兰田径队的比赛中首次亮相。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is brushing a cat\nSentence 2: Nobody is brushing a cat", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nobody is brushing a cat\nSentence 2: A person is brushing a cat", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: MH133\nSentence 2: mh135", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: mh135\nSentence 2: MH133", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天的天气看起来非常好\nSentence 2: 明天的天气看起来飞常好", "response": "-1", "predicted": -1, "true_label": -1}

eval_judge/gpu_results/results_gpu_1.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天的天气看起来飞常好\nSentence 2: 明天的天气看起来非常好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is slicing an eggplant\nSentence 2: A woman is slicing a vegetable", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is slicing a vegetable\nSentence 2: A woman is slicing an eggplant", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么更改来电提示音？\nSentence 2: 如何更改充电提示音？", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 如何更改充电提示音？\nSentence 2: 怎么更改来电提示音？", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The number of extremely obese adults -- those who are at least 100 pounds overweight -- has quadrupled since the 1980s to about 4 million .\nSentence 2: The number of Americans considered extremely obese , or at least 100 pounds overweight , has quadrupled since the 1980s to a startling 4 million , the research shows .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The number of Americans considered extremely obese , or at least 100 pounds overweight , has quadrupled since the 1980s to a startling 4 million , the research shows .\nSentence 2: The number of extremely obese adults -- those who are at least 100 pounds overweight -- has quadrupled since the 1980s to about 4 million .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in glasses and an orange shirt is sitting outside next to a table with drinks on it.\nSentence 2: A man sits near a table with his drinks", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man sits near a table with his drinks\nSentence 2: A man in glasses and an orange shirt is sitting outside next to a table with drinks on it.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 如何让自己的心变得强大呢\nSentence 2: 怎么让自己变得强大", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么让自己变得强大\nSentence 2: 如何让自己的心变得强大呢", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他昨天去商店买了一些苹果的。\nSentence 2: 他昨天去商店买了一些苹。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他昨天去���店买了一些苹。\nSentence 2: 他昨天去商店买了一些苹果的。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is not cooking something\nSentence 2: A woman is cooking eggs", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is cooking eggs\nSentence 2: The woman is not cooking something", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A monkey is brushing the dog\nSentence 2: The monkey is not brushing a bull dog", "response": "3", "predicted": 3, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The monkey is not brushing a bull dog\nSentence 2: A monkey is brushing the dog", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The doctors are reviving a man\nSentence 2: The doctor is helping the patient", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The doctor is helping the patient\nSentence 2: The doctors are reviving a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone is scratching the belly of an animal\nSentence 2: Someone is stroking the belly of an animal", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone is stroking the belly of an animal\nSentence 2: Someone is scratching the belly of an animal", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么黑人喜欢吃白巧克力\nSentence 2: 黑人为什么喜欢吃白巧克力?", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 黑人为什么喜欢吃白巧克力?\nSentence 2: 为什么黑人喜欢吃白巧克力", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two boy are playing on a trampoline\nSentence 2: Nobody is playing on a trampoline", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nobody is playing on a trampoline\nSentence 2: Two boy are playing on a trampoline", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他去商店买了一些苹果和橙子，然而并没有去银行取钱。\nSentence 2: 他去商店买了一些苹果和橙子，然而并没有去把行取钱。", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他去商店买��一些苹果和橙子，然而并没有去把行取钱。\nSentence 2: 他去商店买了一些苹果和橙子，然而并没有去银行取钱。", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 哪里有手工包批发啊？\nSentence 2: 哪里有手工包批发？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 哪里有手工包批发？\nSentence 2: 哪里有手工包批发啊？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 核不扩散条约已有37年的历史。\nSentence 2: 核不扩散条约每5年审查一次。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 核不扩散条约每5年审查一次。\nSentence 2: 核不扩散条约已有37年的历史。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: FIRSTDAY\nSentence 2: Onlyday", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Onlyday\nSentence 2: FIRSTDAY", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is playing a bamboo flute\nSentence 2: The man is not playing a flute", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is not playing a flute\nSentence 2: A man is playing a bamboo flute", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 支付宝支付密码忘记了怎么办啊\nSentence 2: 支付宝支付密码忘记了怎么办", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 支付宝支付密码忘记了怎么办\nSentence 2: 支付宝支付密码忘记了怎么办啊", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他问。以前我回答过了，因为你不爱我。\nSentence 2: 我回答，是因为你不爱我。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我回答，是因为你不爱我。\nSentence 2: 他问。以前我回答过了，因为你不爱我。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 搜索一篇作文\nSentence 2: 植树节是通过什么形式确定的", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 植树节是通过什么形式确定的\nSentence 2: 搜索一篇作文", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Forecasters predict the storm will reach Earth at 3 p.m. eastern time Friday and could last up to 18 hours .\nSentence 2: It is expected to reach Earth about 3 p.m. EDT Friday , and its effects could last 12 to 18 hours .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: It is expected to reach Earth about 3 p.m. EDT Friday , and its effects could last 12 to 18 hours .\nSentence 2: Forecasters predict the storm will reach Earth at 3 p.m. eastern time Friday and could last up to 18 hours .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她紧紧地抱着多罗脱奥说:你为什么要把我弄醒，我的好姐姐啊？\nSentence 2: 她紧紧地抱住多罗特亚，说:我可爱的夫人呀，你为什么要把我叫醒呢？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她紧紧地抱住多罗特亚，说:我可爱的夫人呀，你为什么要把我叫醒呢？\nSentence 2: 她紧紧地抱着多罗脱奥说:你为什么要把我弄醒，我的好姐姐啊？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第28步兵师的第2个BCT集中在拉马迪控制主要道路并保护州长和政府中心。\nSentence 2: 第28步兵师的第二个BCT专注于拉马迪，保护主要道路，控制州长和政府中心。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第28步兵师的第二个BCT专注于拉马迪，保护主要道路，控制州长和政府中心。\nSentence 2: 第28步兵师的第2个BCT集中在拉马迪控制主要道路并保护州长和政府中心。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A golfer is getting ready to putt on the green, with a crowd of people watching in the background.\nSentence 2: The golfer is getting ready to putt on the green, with a crowd watching in the background.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The golfer is getting ready to putt on the green, with a crowd watching in the background.\nSentence 2: A golfer is getting ready to putt on the green, with a crowd of people watching in the background.", "response": "-1", "predicted": -1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man eating some food\nSentence 2: A man is eating some food", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is eating some food\nSentence 2: There is no man eating some food", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 头发怎么变软呢?自己在家就能操作的\nSentence 2: 我的头发偏硬，怎么采能变软", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的头发偏硬，怎么采能变软\nSentence 2: 头发怎么变软呢?自己在家就能操作的", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你脑子进水了吗\nSentence 2: 好吧我爱你", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 好吧我爱你\nSentence 2: 你脑子进水了吗", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在他的妻子去世后，约翰比阿特丽斯与波旁人结婚，他们有两个孩子，卢森堡公爵瓦茨拉夫一世和年轻时去世的波恩。\nSentence 2: 妻子去世后，约翰与波旁的比阿特丽斯结婚。他们有两个孩子，卢森堡公爵Wenceslaus I和年轻时去世的Bonne。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 妻子去世后，约翰与波旁的比阿特丽斯结婚。他们有两个孩子，卢森堡公爵Wenceslaus I和年轻时去世的Bonne。\nSentence 2: 在他的妻子去世后，约翰比阿特丽斯与波旁人结婚，他们有两个孩子，卢森堡公爵瓦茨拉夫一世和年轻时去世的波恩。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An old man looking over a sculpture.\nSentence 2: the man is old", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: the man is old\nSentence 2: An old man looking over a sculpture.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Those conversations had not taken place as of Tuesday night , according to an Oracle spokeswoman .\nSentence 2: Those talks have not taken place , according to an Oracle spokeswoman .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Those talks have not taken place , according to an Oracle spokeswoman .\nSentence 2: Those conversations had not taken place as of Tuesday night , according to an Oracle spokeswoman .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Federal prosecutors , the Securities and Exchange Commission , and the mortgage company 's regulator launched investigations into Freddie Mac 's shake-up .\nSentence 2: The Securities and Exchange Commission and the U.S. Attorney have opened investigations into Freddie Mac over its accounting practices .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The Securities and Exchange Commission and the U.S. Attorney have opened investigations into Freddie Mac over its accounting practices .\nSentence 2: Federal prosecutors , the Securities and Exchange Commission , and the mortgage company 's regulator launched investigations into Freddie Mac 's shake-up .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In a stunning setback for Bill Gates' software empire, the judge in the Microsoft antitrust trial ruled yesterday that the software giant is a monopoly that wielded its power to stifle competition.\nSentence 2: Microsoft holds a monopoly power.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Microsoft holds a monopoly power.\nSentence 2: In a stunning setback for Bill Gates' software empire, the judge in the Microsoft antitrust trial ruled yesterday that the software giant is a monopoly that wielded its power to stifle competition.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: That department, a co-manager of shipwrecks with the other two, has also attempted to address its shipwreck responsibilities since the early '70s.\nSentence 2: Shipwreck salvaging was attempted.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Shipwreck salvaging was attempted.\nSentence 2: That department, a co-manager of shipwrecks with the other two, has also attempted to address its shipwreck responsibilities since the early '70s.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公园散布了。\nSentence 2: 我今天去公园散步聊。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公园散步聊。\nSentence 2: 我今天去公园散布了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: cake\nSentence 2: qcakecakecakecake", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: qcakecakecakecake\nSentence 2: cake", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Retailers J.C. Penney Co . Inc.JCP.N and Walgreen Co.WAG.N kick things off early in the week .\nSentence 2: Retailers J.C. Penney Co . Inc . JCP.N and Walgreen Co . WAG.N kick things off on Monday .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Retailers J.C. Penney Co . Inc . JCP.N and Walgreen Co . WAG.N kick things off on Monday .\nSentence 2: Retailers J.C. Penney Co . Inc.JCP.N and Walgreen Co.WAG.N kick things off early in the week .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么查别人微信记录\nSentence 2: 怎么查别人微信记录？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么查别人微信记录？\nSentence 2: 怎么查别人微信记录", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个男人正系在速记机上。\nSentence 2: 那人用速记。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 那人用速记。\nSentence 2: 一个男人正系在速记机上。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他们的男朋友很有钱吗\nSentence 2: 你很有钱吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你很有钱吗\nSentence 2: 他们的男朋友很有钱吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天的上证指数\nSentence 2: 法拉利多少钱", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 法拉利多少钱\nSentence 2: 今天的上证指数", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head.\nSentence 2: A couple are having their wedding picture taken.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A couple are having their wedding picture taken.\nSentence 2: A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 安逸哟，是什么意思。\nSentence 2: 好安逸是什么意思", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 好安逸是什么意思\nSentence 2: 安逸哟，是什么意思。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么？因为在您和她的关系问题没有明确解决以前，她不好作出选择从我这方面说，问题已经得到明确的解决。\nSentence 2: 因为在您跟她的关系问题没有最后解决以前，她不能做出其他任何抉择。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 因为在您跟她的关系问题没有最后解决以前，她不能做出其他任何抉择。\nSentence 2: 为什么？因为在您和她的关系问题没有明确解决以前，她不好作出选择从我这方面说，问题已经得到明确的解决。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In addition , the Justice Department said that the FBI has conducted ' ' fewer than 10 ' ' investigations involving visits to mosques .\nSentence 2: In addition , \" fewer than 10 \" FBI offices have conducted investigations involving visits to Islamic mosques , the Justice Department said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In addition , \" fewer than 10 \" FBI offices have conducted investigations involving visits to Islamic mosques , the Justice Department said .\nSentence 2: In addition , the Justice Department said that the FBI has conducted ' ' fewer than 10 ' ' investigations involving visits to mosques .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Drinking four or five cups of green tea a day can help protect against cancer and heart disease, aid in weight loss, ease arthritis and chronic pain and even help prevent Alzheimer's disease.\nSentence 2: Tea protects from some diseases.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tea protects from some diseases.\nSentence 2: Drinking four or five cups of green tea a day can help protect against cancer and heart disease, aid in weight loss, ease arthritis and chronic pain and even help prevent Alzheimer's disease.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is pouring oil into a skillet\nSentence 2: A man is taking the oil out off a skillet", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is taking the oil out off a skillet\nSentence 2: A man is pouring oil into a skillet", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 奥迪A四\nSentence 2: 宝马五系和奥迪A4L有什么区别", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 宝马五系和奥迪A4L有什么区别\nSentence 2: 奥迪A四", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman squeezing juice out of a lemon\nSentence 2: A woman is squeezing a lemon", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is squeezing a lemon\nSentence 2: There is no woman squeezing juice out of a lemon", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 播放一首桥边姑娘\nSentence 2: 播放酷酷狗音乐", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 播放酷酷狗音乐\nSentence 2: 播放一首桥边姑娘", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: On August 15, IDFA submitted its comments regarding the U.S. Department of Agriculture's (USDA) latest draft of specifications (known as \"Dairy 6\") for government purchases of salted butter, cheddar cheese (blocks), American-style cheese (barrels) and nonfat dry milk.\nSentence 2: A government purchases food.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A government purchases food.\nSentence 2: On August 15, IDFA submitted its comments regarding the U.S. Department of Agriculture's (USDA) latest draft of specifications (known as \"Dairy 6\") for government purchases of salted butter, cheddar cheese (blocks), American-style cheese (barrels) and nonfat dry milk.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An old woman is shaking hands with a man\nSentence 2: A man and a woman are shaking hands", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are shaking hands\nSentence 2: An old woman is shaking hands with a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 缝的时候，他两手发抖，但是尽力克制住，缝上以后，他又把大衣穿上，。\nSentence 2: 他缝的时候两手发抖，但他克制住了，他缝得很好，再穿上大衣的时候，从外边一点也看不出来。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他缝的时候两手发抖，但他克制住了，他缝得很好，再穿上大衣的时候，从外边一点也看不出来。\nSentence 2: 缝的时候，他两手发抖，但是尽力克制住，缝上以后，他又把大衣穿上，。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Of the over 400 different species of sharks found in the world's oceans, only about 30 have been reported to ever attack a human.\nSentence 2: A shark attacked a human being.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A shark attacked a human being.\nSentence 2: Of the over 400 different species of sharks found in the world's oceans, only about 30 have been reported to ever attack a human.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The herd of caribous is crossing a road\nSentence 2: The herd of caribous is not crossing a road", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The herd of caribous is not crossing a road\nSentence 2: The herd of caribous is crossing a road", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A study of more than 90,000 Japanese found that people who drank coffee daily or nearly every day had half the liver cancer risk of those who never drank coffee. The protective effect occurred in people who drank one to two cups a day and increased at three to four cups.\nSentence 2: Tea protects from some diseases.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tea protects from some diseases.\nSentence 2: A study of more than 90,000 Japanese found that people who drank coffee daily or nearly every day had half the liver cancer risk of those who never drank coffee. The protective effect occurred in people who drank one to two cups a day and increased at three to four cups.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打开相机设置\nSentence 2: 照相机打开相机", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 照相机打开相机\nSentence 2: 打开相机设置", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A head-on collision between a Dhaka bound bus and a motorcycle left three people dead at Gangabardi area on the Faridpur-Khulna Highway on September 26.\nSentence 2: People were killed in a collision between ships.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People were killed in a collision between ships.\nSentence 2: A head-on collision between a Dhaka bound bus and a motorcycle left three people dead at Gangabardi area on the Faridpur-Khulna Highway on September 26.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天的天气真不错\nSentence 2: 今天的天气真不策", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天的天气真不策\nSentence 2: 今天的天气真不错", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个男人在玩扑克牌。\nSentence 2: 一个男人在玩纸牌戏法。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个男人在玩纸牌戏法。\nSentence 2: 一个男人在玩扑克牌。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 各种标志的含义\nSentence 2: 这些标志的含义", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这些标志的含义\nSentence 2: 各种标志的含义", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Even today, within the deepest recesses of our mind, lies a primordial fear that will not allow us to enter the sea without thinking about the possibility of being attacked by a shark.\nSentence 2: A shark attacked a human being.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A shark attacked a human being.\nSentence 2: Even today, within the deepest recesses of our mind, lies a primordial fear that will not allow us to enter the sea without thinking about the possibility of being attacked by a shark.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么中国校服那么难看？\nSentence 2: 中国校服为什么那么难看？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国校服为什么那么难看？\nSentence 2: 为什么中国校服那么难看？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two men spar in martial arts gear.\nSentence 2: There are multiple people present.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are multiple people present.\nSentence 2: Two men spar in martial arts gear.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nobody is riding a bike\nSentence 2: Two people are riding a bike", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people are riding a bike\nSentence 2: Nobody is riding a bike", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气很好\nSentence 2: 我今天天天气很好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天天气很好\nSentence 2: 我今天天气很好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 语文的学习方法\nSentence 2: 学习语文的方法", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 学习语文的方法\nSentence 2: 语文的学习方法", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 腰酸背痛，而且很饿怎么回事\nSentence 2: 腰酸背痛怎么回事啊", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 腰酸背痛怎么回事啊\nSentence 2: 腰酸背痛，而且很饿怎么回事", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: “以马内利”是什么意思？\nSentence 2: 以马内利是什么意思?", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 以马内利是什么意思?\nSentence 2: “以马内利”是什么意思？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The military officer is barking at the recruits\nSentence 2: An officer is talking to the recruits", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An officer is talking to the recruits\nSentence 2: The military officer is barking at the recruits", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 温州售后电话\nSentence 2: 电话放入电脑传送东西没回应", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 电话放入电脑传送东西没回应\nSentence 2: 温州售后电话", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我妹妹是谁\nSentence 2: 你知道我的妹妹叫什么名字吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你知道我的妹妹叫什么名字吗\nSentence 2: 我妹妹是谁", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is eating a pizza\nSentence 2: A man is eating some food", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is eating some food\nSentence 2: A man is eating a pizza", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three men are sitting outside on chairs with red seats.\nSentence 2: Men are sitting outside.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Men are sitting outside.\nSentence 2: Three men are sitting outside on chairs with red seats.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 沙特阿拉伯的死罪包括强奸、谋杀、叛教、武装抢劫和贩毒。\nSentence 2: 伊朗的死罪包括谋杀、强奸、武装抢劫、严重贩毒和通奸。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 伊朗的死罪包括谋杀、强奸、武装抢劫、严重贩毒和通奸。\nSentence 2: 沙特阿拉伯的死罪包括强奸、谋杀、叛教、武装抢劫和贩毒。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a30\nSentence 2: OPPOA33", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOA33\nSentence 2: a30", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样制作表格\nSentence 2: 怎样学表格制作", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样学表格制作\nSentence 2: 怎样制作表格", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天气很好，我们可以去散步。\nSentence 2: 天气很好，我们可已去散步。", "response": "5", "predicted": 5, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天气很好，我们可已去散步。\nSentence 2: 天气很好，我们可以去散步。", "response": "5", "predicted": 5, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气很好\nSentence 2: 我今天天气很好好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气很好好\nSentence 2: 我今天天气很好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man with no shirt on is performing with a baton.\nSentence 2: A man is doing things with a baton.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is doing things with a baton.\nSentence 2: A man with no shirt on is performing with a baton.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: December\nSentence 2: February", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: February\nSentence 2: December", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child in a ninja outfit does a jumping kick.\nSentence 2: a child does a jumping kick", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a child does a jumping kick\nSentence 2: A child in a ninja outfit does a jumping kick.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 语音助手无法语音解锁怎么办\nSentence 2: 语音助手无法开启", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 语音助手无法开启\nSentence 2: 语音助手无法语音解锁怎么办", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is driving a car\nSentence 2: A man is driving a vehicle", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is driving a vehicle\nSentence 2: A man is driving a car", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 从你的全世界路过\nSentence 2: 从你的全世界里路过", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 从你的全世界里路过\nSentence 2: 从你的全世界路过", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天是什么日子\nSentence 2: 第七次全国人口普查摸底工作时间节点为", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第七次全国人口普查摸底工作时间节点为\nSentence 2: 明天是什么日子", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 癌前病变明显是典型的组织，在显微检镜查中看起来异常，并且相比其在形态上正常的对应物，癌症更有可能出现在其中。\nSentence 2: 癌前病变明显是一种典型组织，在用显微镜检查时其会呈现异常，而且在此处发生癌变的可能性大于其形态正常的对应组织。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 癌前病变明显是一种典型组织，在用显微镜检查时其会呈现异常，而且在此处发生癌变的可能性大于其形态正常的对应组织。\nSentence 2: 癌前病变明显是典型的组织，在显微检镜查中看起来异常，并且相比其在形态上正常的对应物，癌症更有可能出现在其中。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 李现生日快乐\nSentence 2: 王者荣耀刘备", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 王者荣耀刘备\nSentence 2: 李现生日快乐", "response": "0", "predicted": 0, "true_label": 0}

eval_judge/gpu_results/results_gpu_2.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天去看了一场电影，电影非常有趣。\nSentence 2: 我昨天去看了一场电影，电影非常有去。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天去看了一场电影，电影非常有去。\nSentence 2: 我昨天去看了一场电影，电影非常有趣。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我去散步了。\nSentence 2: 今天天气真好，我去散步了。去 散布。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我去散步了。去 散布。\nSentence 2: 今天天气真好，我去散步了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜欢吃苹郭果\nSentence 2: 苹果是我喜欢的水果", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 苹果是我喜欢的水果\nSentence 2: 我喜欢吃苹郭果", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: ILOVEYOUCHINA\nSentence 2: MadeinChina", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: MadeinChina\nSentence 2: ILOVEYOUCHINA", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃灯笼椒吗\nSentence 2: 橘子有什么好处", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 橘子有什么好处\nSentence 2: 怀孕能吃灯笼椒吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去超市买了些蔬菜。\nSentence 2: 我今天天超市买了些蔬菜。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天超市买了些蔬菜。\nSentence 2: 我今天去超市买了些蔬菜。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: These folk art traditions have been preserved for hundreds of years.\nSentence 2: Indigenous folk art is preserved.", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Indigenous folk art is preserved.\nSentence 2: These folk art traditions have been preserved for hundreds of years.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 李白写了多少诗\nSentence 2: 笑问客从何处来的上一句", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 笑问客从何处来的上一句\nSentence 2: 李白写了多少诗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这两个人即将在水球锦标赛中得分。\nSentence 2: 这两个人将要在一场比赛中竞争。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这两个人将要在一场比赛中竞争。\nSentence 2: 这两个人即将在水球锦标赛中得分。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: LETSGO\nSentence 2: Letmego", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Letmego\nSentence 2: LETSGO", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is waking up\nSentence 2: A child is waking up", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child is waking up\nSentence 2: A girl is waking up", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 其他欧洲大国不再准备接受新的法国扩张，并愿意组建联盟抵制这种事情。\nSentence 2: 其他欧洲大国不再接受新的法国扩张，并准备组成联盟反对这样的事情。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 其他欧洲大国不再接受新的法国扩张，并准备组成联盟反对这样的事情。\nSentence 2: 其他欧洲大国不再准备接受新的法国扩张，并愿意组建联盟抵制这种事情。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我随时都在等待着您的，也就是说，等待令堂和令妹到来拉斯科利尼科夫稍动了动，想说什么；他的脸上露出激动不安的神情。\nSentence 2: 拉斯柯尔尼科夫动了动，好像要开口说话，他脸上现出略为激动的样子。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 拉斯柯尔尼科夫动了动，好像要开口说话，他脸上现出略为激动的样子。\nSentence 2: 我随时都在等待着您的，也就是说，等待令堂和令妹到来拉斯科利尼科夫稍动了动，想说什么；他的脸上露出激动不安的神情。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is cutting cooked octopus\nSentence 2: A woman is cutting an octopus", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is cutting an octopus\nSentence 2: The woman is cutting cooked octopus", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: CLASSROOM\nSentence 2: Whatsinyourclassroom", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whatsinyourclassroom\nSentence 2: CLASSROOM", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young boy in a cowboy hat rides a large horse.\nSentence 2: A boy is pretending to be a cowboy.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy is pretending to be a cowboy.\nSentence 2: A young boy in a cowboy hat rides a large horse.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这个是什么水果\nSentence 2: 请问，这是什么水果？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问，这是什么水果？\nSentence 2: 请问这个是什么水果", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: lsquo\nSentence 2: oppor9splusota", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor9splusota\nSentence 2: lsquo", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman in black reviews a message as she walks to work.\nSentence 2: The woman in black has a job.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman in black has a job.\nSentence 2: A woman in black reviews a message as she walks to work.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真不索\nSentence 2: 今天天气真好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好\nSentence 2: 今天天气真不索", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man with a hard hat dancing\nSentence 2: A man with a hard hat is dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man with a hard hat is dancing\nSentence 2: There is no man with a hard hat dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他取代约翰马布里作为德里克李的第一垒的替补。\nSentence 2: 他取代约翰马布里作为第一基地的替补德里克李。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他取代约翰马布里作为第一基地的替补德里克李。\nSentence 2: 他取代约翰马布里作为德里克李的第一垒的替补。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国最高政治顾问会见希腊总理\nSentence 2: 中国高级政治顾问会见也门总统", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国高级政治顾问会见也门总统\nSentence 2: 中国最高政治顾问会见希腊总理", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Antiemetic drugs are drugs used to combat nausea and vomiting.\nSentence 2: Drugs are used to combat schizophrenia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Drugs are used to combat schizophrenia.\nSentence 2: Antiemetic drugs are drugs used to combat nausea and vomiting.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气很漂亮\nSentence 2: 我今天天气很漂流", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气很漂流\nSentence 2: 我今天天气很漂亮", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 人民币兑美元汇率周一跌至6.3282。\nSentence 2: 周四人民币兑美元汇率回落至6.2697。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 周四人民币兑美元汇率回落至6.2697。\nSentence 2: 人民币兑美元汇率周一跌至6.3282。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man riding a mechanical bull\nSentence 2: The man is riding a mechanical bull", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is riding a mechanical bull\nSentence 2: There is no man riding a mechanical bull", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 十亿。世界上的每一个人都达到了墨西哥目前的繁荣水平。\nSentence 2: 世界上的每一个人都处在目前西北非洲的“繁荣”水平。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 世界上的每一个人都处在目前西北非洲的“繁荣”水平。\nSentence 2: 十亿。世界上的每一个人都达到了墨西哥目前的繁荣水平。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl in a black sweater and jeans pours water into an empty soda bottle.\nSentence 2: Someone pours water into a bottle", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone pours water into a bottle\nSentence 2: A girl in a black sweater and jeans pours water into an empty soda bottle.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: State Supreme Court Justice Ira Gammerman said in court this morning , \" It was an ill-conceived lawsuit . \"\nSentence 2: New York Supreme Court Justice Ira Gammerman said in his statement that the lawsuit was \" ill-conceived . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: New York Supreme Court Justice Ira Gammerman said in his statement that the lawsuit was \" ill-conceived . \"\nSentence 2: State Supreme Court Justice Ira Gammerman said in court this morning , \" It was an ill-conceived lawsuit . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man draws a caricature of a woman.\nSentence 2: A man drawing a caricature of his wife.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man drawing a caricature of his wife.\nSentence 2: A man draws a caricature of a woman.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这种植物叫什么名字\nSentence 2: 请问这种植物叫什么名字、", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这种植物叫什么名字、\nSentence 2: 请问这种植物叫什么名字", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 官方SpA新闻社发表的一份声明称，al-Dosari被判犯有贩毒罪。\nSentence 2: 强奸犯被确认为HosseinSafarlou，并被判定犯有贩毒罪。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 强奸犯被确认为HosseinSafarlou，并被判定犯有贩毒罪。\nSentence 2: 官方SpA新闻社发表的一份声明称，al-Dosari被判犯有贩毒罪。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Aki Kaurismaki was born in Orimattila, Finland, on 4th April, 1957. After studying film, he wrote, produced and directed Saimaailmio (1981) in Finland.\nSentence 2: Aki Kaurismaki directed a film.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Aki Kaurismaki directed a film.\nSentence 2: Aki Kaurismaki was born in Orimattila, Finland, on 4th April, 1957. After studying film, he wrote, produced and directed Saimaailmio (1981) in Finland.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 关于友谊的名人名言\nSentence 2: 有关天气的谚语", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有关天气的谚语\nSentence 2: 关于友谊的名人名言", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你们现在都在干嘛。\nSentence 2: 现在都在干嘛呢", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 现在都在干嘛呢\nSentence 2: 你们现在都在干嘛。", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no young boy playing a wind instrument\nSentence 2: A young boy is playing a wind instrument", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young boy is playing a wind instrument\nSentence 2: There is no young boy playing a wind instrument", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Did the media pounce upon the religion of Jurgen Schneider, the developer who went bankrupt and fled Germany owing billions of DMs to Deutsche Bank and others, as the cause of his downfall?\nSentence 2: Jurgen Schneider went bankrupt.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Jurgen Schneider went bankrupt.\nSentence 2: Did the media pounce upon the religion of Jurgen Schneider, the developer who went bankrupt and fled Germany owing billions of DMs to Deutsche Bank and others, as the cause of his downfall?", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你挺好的啊\nSentence 2: 你挺能干的话", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你挺能干的话\nSentence 2: 你挺好的啊", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Ahold Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for the scandal .\nSentence 2: Board Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for events at Foodservice and elsewhere in the embattled group .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Board Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for events at Foodservice and elsewhere in the embattled group .\nSentence 2: Ahold Chairman Henny de Ruiter offered shareholders at the meeting his \" sincere apologies \" for the scandal .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A picture is being drawn by a man\nSentence 2: The person is drawing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The person is drawing\nSentence 2: A picture is being drawn by a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three dogs running through a field.\nSentence 2: Three dogs are running", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three dogs are running\nSentence 2: Three dogs running through a field.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man with a paddle stands next to a woman with a bikini top; and her arms are in a dance pose.\nSentence 2: someone dances while a man holds a paddle", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: someone dances while a man holds a paddle\nSentence 2: A man with a paddle stands next to a woman with a bikini top; and her arms are in a dance pose.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我想玩抖音\nSentence 2: 继续播放抖音", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 继续播放抖音\nSentence 2: 我想玩抖音", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is wearing a hard hat and dancing\nSentence 2: There is no man with a hard hat dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man with a hard hat dancing\nSentence 2: A man is wearing a hard hat and dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The CFAP purchases food stamp coupons from the federal government and distributes them to eligible recipients.\nSentence 2: A government purchases food.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A government purchases food.\nSentence 2: The CFAP purchases food stamp coupons from the federal government and distributes them to eligible recipients.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 北京哪里租房子最便宜？\nSentence 2: 北京哪里租房子最便宜？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 北京哪里租房子最便宜？\nSentence 2: 北京哪里租房子最便宜？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我想给我的饭店起个名字\nSentence 2: 我想给小饭店起个名字，", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我想给小饭店起个名字，\nSentence 2: 我想给我的饭店起个名字", "response": "-1", "predicted": -1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在狭窄的小路上骑自行车的人。\nSentence 2: 一名男子骑着独轮车沿着一条铺好的道路行驶。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一名男子骑着独轮车沿着一条铺好的道路行驶。\nSentence 2: 在狭窄的小路上骑自行车的人。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is playing the guitar\nSentence 2: Nobody is playing the guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nobody is playing the guitar\nSentence 2: A man is playing the guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天昨了个美梦\nSentence 2: 我昨天做了个美梦", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天做了个美梦\nSentence 2: 我昨天昨了个美梦", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天会下雨吗？\nSentence 2: 明天会下雨马？", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天会下雨马？\nSentence 2: 明天会下雨吗？", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tony winners will be announced June 8 at the Radio City Music Hall in New York .\nSentence 2: Winners will be announced in a June 8 ceremony broadcast on CBS from Radio City Music Hall .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Winners will be announced in a June 8 ceremony broadcast on CBS from Radio City Music Hall .\nSentence 2: Tony winners will be announced June 8 at the Radio City Music Hall in New York .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The animal with big eyes is voraciously eating\nSentence 2: The animal with big eyes is eating", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The animal with big eyes is eating\nSentence 2: The animal with big eyes is voraciously eating", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A female runner dressed in blue athletic wear is running in a competition, while spectators line the street.\nSentence 2: There are people on the street.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are people on the street.\nSentence 2: A female runner dressed in blue athletic wear is running in a competition, while spectators line the street.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: whatyou\nSentence 2: withoutyou", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: withoutyou\nSentence 2: whatyou", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman playing an electric guitar\nSentence 2: A woman is playing an electric guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is playing an electric guitar\nSentence 2: There is no woman playing an electric guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 呼叫爸爸呼叫爸爸\nSentence 2: 呼叫起来", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 呼叫起来\nSentence 2: 呼叫爸爸呼叫爸爸", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 求这个动态图的出处\nSentence 2: 动态图，求出处。", "response": "4", "predicted": 4, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 动态图，求出处。\nSentence 2: 求这个动态图的出处", "response": "动态图，求出处。", "predicted": -100, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这完全取决于你的公司和你的合同。\nSentence 2: 那要看你的公司了。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 那要看你的公司了。\nSentence 2: 这完全取决于你的公司和你的合同。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第三方理财公司可信吗?\nSentence 2: 投资理财公司可信吗？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 投资理财公司可信吗？\nSentence 2: 第三方理财公司可信吗?", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoR15\nSentence 2: oppr15", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppr15\nSentence 2: oppoR15", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The petition alleges that Huletts unfair sales have damaged the U.S. industry , sending market prices below sustainable levels .\nSentence 2: Those unfair sales have damaged the US industry by eroding market prices below sustainable levels , says Alcoa .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Those unfair sales have damaged the US industry by eroding market prices below sustainable levels , says Alcoa .\nSentence 2: The petition alleges that Huletts unfair sales have damaged the U.S. industry , sending market prices below sustainable levels .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: READING\nSentence 2: REVIEW", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: REVIEW\nSentence 2: READING", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young child joyfully pulls colorful tissue paper from a decorated box, looking for his present.\nSentence 2: a child pulls colorful tissue paper from a fancy box", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a child pulls colorful tissue paper from a fancy box\nSentence 2: A young child joyfully pulls colorful tissue paper from a decorated box, looking for his present.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 五音有什么\nSentence 2: 零点二五等于多少", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 零点二五等于多少\nSentence 2: 五音有什么", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样调理痛经呢？\nSentence 2: 痛经怎么调理，吃什么好？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 痛经怎么调理，吃什么好？\nSentence 2: 怎样调理痛经呢？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 帮我打开美颜自拍\nSentence 2: 给我拍照", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 给我拍照\nSentence 2: 帮我打开美颜自拍", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Boy in midair on bicycle\nSentence 2: A boy is airborne on a bicycle.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy is airborne on a bicycle.\nSentence 2: Boy in midair on bicycle", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 伊朗近几个月来明显增加了处决人数，因为伊朗官员正在实施一项旨在促进道德和加强伊朗安全的运动。\nSentence 2: 伊朗官员在2007年增加了处决人数，因为伊朗官员实施了一场据称旨在促进道德和加强社会安全的运动。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 伊朗官员在2007年增加了处决人数，因为伊朗官员实施了一场据称旨在促进道德和加强社会安全的运动。\nSentence 2: 伊朗近几个月来明显增加了处决人数，因为伊朗官员正在实施一项旨在促进道德和加强伊朗安全的运动。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么软件看电影最好。\nSentence 2: 看电影用什么软件最好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 看电影用什么软件最好\nSentence 2: 什么软件看电影最好。", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。\nSentence 2: 音乐由A. T. Ummer编写，歌词由Koorkkancheri Sugathan和Poovachal Khader创作。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 音乐由A. T. Ummer编写，歌词由Koorkkancheri Sugathan和Poovachal Khader创作。\nSentence 2: 音乐由A. T. Ummer创作，歌词由Koorkkancheri Sugathan和Poovachal Khader编写。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is slowly dancing on the road\nSentence 2: A man is dancing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is dancing\nSentence 2: A man is slowly dancing on the road", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Man jumping up in the air and splitting his legs with his mouth open.\nSentence 2: A man is jumping.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is jumping.\nSentence 2: Man jumping up in the air and splitting his legs with his mouth open.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Iran has yet to sign an additional protocol to the NPT treaty which would allow U.N. inspections at short notice .\nSentence 2: Iran has yet to sign an additional protocol to the Nuclear Non-Proliferation Treaty , which it signed in 1970 , that would allow IAEA inspections at short notice .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Iran has yet to sign an additional protocol to the Nuclear Non-Proliferation Treaty , which it signed in 1970 , that would allow IAEA inspections at short notice .\nSentence 2: Iran has yet to sign an additional protocol to the NPT treaty which would allow U.N. inspections at short notice .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样开通农行手机银行\nSentence 2: 怎样开通农行手机银行？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样开通农行手机银行？\nSentence 2: 怎样开通农行手机银行", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Soccer players are kicking a soccer ball into the goal\nSentence 2: There are no soccer players kicking a soccer ball into the goal", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are no soccer players kicking a soccer ball into the goal\nSentence 2: Soccer players are kicking a soccer ball into the goal", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国联通中国移动中国电信哪个好?\nSentence 2: 中国电信中国移动中国联通上网哪个好", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国电信中国移动中国联通上网哪个好\nSentence 2: 中国联通中国移动中国电信哪个好?", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 来自世界各地的编辑敦促土耳其保护媒体\nSentence 2: 土耳其打击反政府武装，媒体自由岌岌可危", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 土耳其打击反政府武装，媒体自由岌岌可危\nSentence 2: 来自世界各地的编辑敦促土耳其保护媒体", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: This model allows for different levels of training to be offered to the various skill levels of soldiers that will be operating and maintaining a particular system.\nSentence 2: Soldiers are trained using virtual reality.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Soldiers are trained using virtual reality.\nSentence 2: This model allows for different levels of training to be offered to the various skill levels of soldiers that will be operating and maintaining a particular system.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone is drawing a person\nSentence 2: The person is drawing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The person is drawing\nSentence 2: Someone is drawing a person", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 吃完螃蟹能吃橘子吗\nSentence 2: 鱼吃什么", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 鱼吃什么\nSentence 2: 吃完螃蟹能吃橘子吗", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 课一结束，骚动便随之而来，但她的话音刚落，全校又复归平静，她继续说:今天早晨的早饭，你们都吃不下去，大家一定饿坏了，我己经吩咐给大家准备了面包和乳酪当点心，教师们带着某种惊异的目光看着她。\nSentence 2: 她接着往下说:今天早上你们早饭吃不下去；现在一定都饿了；——我已经吩咐过，给大伙儿准备一顿面包和干酪的点心。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她接着往下说:今天早上你们早饭吃不下去；现在一定都饿了；——我已经吩咐过，给大伙儿准备一顿面包和干酪的点心。\nSentence 2: 课一结束，骚动便随之而来，但她的话音刚落，全校又复归平静，她继续说:今天早晨的早饭，你们都吃不下去，大家一定饿坏了，我己经吩咐给大家准备了面包和乳酪当点心，教师们带着某种惊异的目光看着她。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An woman that is old is shaking hands with a man\nSentence 2: A man and a woman are shaking hands", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are shaking hands\nSentence 2: An woman that is old is shaking hands with a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Man holding torch of fire.\nSentence 2: There is a man.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is a man.\nSentence 2: Man holding torch of fire.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is applying makeup to her face\nSentence 2: There is no woman putting on makeup", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman putting on makeup\nSentence 2: A girl is applying makeup to her face", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In the early 20th century, agoraphobia, or the fear of public spaces, saw a transformation in its demographic composition. The disease's early history can be characterized as a predominantly male disorder where the majority of patients diagnosed with the disease were white, middle-class men.\nSentence 2: Agoraphobia is a widespread disorder.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Agoraphobia is a widespread disorder.\nSentence 2: In the early 20th century, agoraphobia, or the fear of public spaces, saw a transformation in its demographic composition. The disease's early history can be characterized as a predominantly male disorder where the majority of patients diagnosed with the disease were white, middle-class men.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 赛普拉斯县由MHCW的联邦选举部门服务，并由保守党议员GLEN MOTZ代表加拿大下议院。\nSentence 2: 赛普拉斯县由MHCW的联邦选举部门代表，并由保守党成员GLEN MOTZ在加拿大下议院任职。", "response": "5", "predicted": 5, "true_label": 4}

eval_judge/gpu_results/results_gpu_3.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 赛普拉斯县由MHCW的联邦选举部门代表，并由保守党成员GLEN MOTZ在加拿大下议院任职。\nSentence 2: 赛普拉斯县由MHCW的联邦选举部门服务，并由保守党议员GLEN MOTZ代表加拿大下议院。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机怎么扩展运行内存\nSentence 2: 怎么增加手机运行内存", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么增加手机运行内存\nSentence 2: 手机怎么扩展运行内存", "response": "1", "predicted": 1, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two dirt bike riders, one wearing green and the other wearing blue and white, are jumping a hill.\nSentence 2: Two dirt bike riders are outside", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two dirt bike riders are outside\nSentence 2: Two dirt bike riders, one wearing green and the other wearing blue and white, are jumping a hill.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The recent collision between a Super Ferry and a smaller ship that killed at least 38 people has once again put the spotlight on the Philippine Coast Guard.\nSentence 2: People were killed in a collision between ships.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People were killed in a collision between ships.\nSentence 2: The recent collision between a Super Ferry and a smaller ship that killed at least 38 people has once again put the spotlight on the Philippine Coast Guard.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is phoning\nSentence 2: A man is talking on the phone", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is talking on the phone\nSentence 2: A man is phoning", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 很酷的动漫女孩图片\nSentence 2: 动漫可爱女孩的图片", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 动漫可爱女孩的图片\nSentence 2: 很酷的动漫女孩图片", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人在割花。\nSentence 2: 一个女人在切西兰花。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人在切西兰花。\nSentence 2: 一个女人在割花。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 甲沟炎怎么在家自己治疗\nSentence 2: 甲沟炎怎么治疗【甲沟炎】", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 甲沟炎怎么治疗【甲沟炎】\nSentence 2: 甲沟炎怎么在家自己治疗", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了苹果和\nSentence 2: 今天早上吃了苹果和牛奶", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天早上吃了苹果和牛奶\nSentence 2: 我今天吃了苹果和", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气晴朗，我去公园散步。\nSentence 2: 今天天气晴朗，我去公园伞步。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气晴朗，我去公园伞步。\nSentence 2: 今天天气晴朗，我去公园散步。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 觉得生活没意思咋办\nSentence 2: 有时候感觉生活真没意思", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有时候感觉生活真没意思\nSentence 2: 觉得生活没意思咋办", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 被斯蒂芬·曼根（Stephen Mangan）阅读的佩尼·莱斯特（Penny Leicester）删除，彼得·塞拉芬诺维奇（Peter Serafinowicz）作为“指南”（The Guide）的声音，由希瑟·拉莫尔（Heather Larmour）制作。\nSentence 2: 它由Penny Leicester缩短，由Stephen Mangan阅读，由Peter Serafinowicz作为导游的声音和Heather Larmour制作。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它由Penny Leicester缩短，由Stephen Mangan阅读，由Peter Serafinowicz作为导游的声音和Heather Larmour制作。\nSentence 2: 被斯蒂芬·曼根（Stephen Mangan）阅读的佩尼·莱斯特（Penny Leicester）删除，彼得·塞拉芬诺维奇（Peter Serafinowicz）作为“指南”（The Guide）的声音，由希瑟·拉莫尔（Heather Larmour）制作。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPO的设计理念是\nSentence 2: 善良都是个性的正面", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 善良都是个性的正面\nSentence 2: OPPO的设计理念是", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The young boy in a red shirt and jeans resists his sister's request for water on the road.\nSentence 2: Kids don't share water.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Kids don't share water.\nSentence 2: The young boy in a red shirt and jeans resists his sister's request for water on the road.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoA\nSentence 2: oppoa79ota", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoa79ota\nSentence 2: oppoA", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people play horseshoes.\nSentence 2: Tom and Sue were playing outside.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tom and Sue were playing outside.\nSentence 2: Two people play horseshoes.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 求现代军事题材电视剧\nSentence 2: 国产现代军事题材的电视剧", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 国产现代军事题材的电视剧\nSentence 2: 求现代军事题材电视剧", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The United States has threatened to end U.N. peacekeeping in Bosnia unless American personnel are given immunity from prosecution by the new international war-crimes tribunal.\nSentence 2: U.N. peacekeeping forces withdrew from Bosnia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: U.N. peacekeeping forces withdrew from Bosnia.\nSentence 2: The United States has threatened to end U.N. peacekeeping in Bosnia unless American personnel are given immunity from prosecution by the new international war-crimes tribunal.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two men are fighting\nSentence 2: Some men are fighting", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some men are fighting\nSentence 2: Two men are fighting", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 主屏幕时间怎样没有显示天气\nSentence 2: 怎样在屏幕上显示时间和天气预报", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样在屏幕上显示时间和天气预报\nSentence 2: 主屏幕时间怎样没有显示天气", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cloud9是BeagleBone黑色单板计算机的原生IDE，主要在node.js的扩展中编程，称为Bonescript。\nSentence 2: Cloud9是本机板载计算机BeagleBone Black的唯一IDE，它主要在node.js的扩展中编程，称为Bonescript。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cloud9是本机板载计算机BeagleBone Black的唯一IDE，它主要在node.js的扩展中编程，称为Bonescript。\nSentence 2: Cloud9是BeagleBone黑色单板计算机的原生IDE，主要在node.js的扩展中编程，称为Bonescript。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 设置输入法皮肤设置输入法皮肤\nSentence 2: 输入法设置", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 输入法设置\nSentence 2: 设置输入法皮肤设置输入法皮肤", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 丑是我的代名词。\nSentence 2: 丑是儿子的代名词。", "response": "2", "predicted": 2, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 丑是儿子的代名词。\nSentence 2: 丑是我的代名词。", "response": "2", "predicted": 2, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 通过两个通道上支持的数据，可以复制冗余通信。\nSentence 2: 两个通道上的复制数据都支持冗余通信。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两个通道上的复制数据都支持冗余通信。\nSentence 2: 通过两个通道上支持的数据，可以复制冗余通信。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的手机卡开关开不了\nSentence 2: 我的手机卡有流量吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的手机卡有流量吗\nSentence 2: 我的手机卡开关开不了", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在那里，Anna Wallner遇到了Matisic并创造了“购物袋”概念。\nSentence 2: 在那里，Matisic遇到了Anna Wallner并创造了“购物袋”的概念。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在那里，Matisic遇到了Anna Wallner并创造了“购物袋”的概念。\nSentence 2: 在那里，Anna Wallner遇到了Matisic并创造了“购物袋”概念。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Yet another big fight has been over a White House proposal to privatize air-traffic control at major airports .\nSentence 2: Yet another fight has been waged over a White House proposal to privatize air traffic control at major airports , such as those in Pittsburgh and Philadelphia .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Yet another fight has been waged over a White House proposal to privatize air traffic control at major airports , such as those in Pittsburgh and Philadelphia .\nSentence 2: Yet another big fight has been over a White House proposal to privatize air-traffic control at major airports .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 人生得意须尽欢下一句是啥\nSentence 2: 人生是啥", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 人生是啥\nSentence 2: 人生得意须尽欢下一句是啥", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 纽约证券交易所(NewYorkStockExchange)下跌股与上涨股之比接近2：1。\nSentence 2: 纽约证交所上涨股与下跌股之比接近8：3，纳斯达克(Nasdaq)上涨股与下跌股之比超过11：5。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 纽约证交所上涨股与下跌股之比接近8：3，纳斯达克(Nasdaq)上涨股与下跌股之比超过11：5。\nSentence 2: 纽约证券交易所(NewYorkStockExchange)下跌股与上涨股之比接近2：1。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man on stage is singing into the microphone\nSentence 2: The man on stage isn 't singing into the microphone", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man on stage isn 't singing into the microphone\nSentence 2: The man on stage is singing into the microphone", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Defense lawyers had objected in pretrial hearings to the videotape and the photo , saying they were too unclear to identify Muhammad .\nSentence 2: Defense lawyers had objected in earlier hearings to showing the videotape and the photo , saying they were too unclear to identify the person in them .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Defense lawyers had objected in earlier hearings to showing the videotape and the photo , saying they were too unclear to identify the person in them .\nSentence 2: Defense lawyers had objected in pretrial hearings to the videotape and the photo , saying they were too unclear to identify Muhammad .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of children in African clothing.\nSentence 2: A person in clothing", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person in clothing\nSentence 2: A group of children in African clothing.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Although it 's unclear whether Sobig was to blame , The New York Times also asked employees at its headquarters yesterday to shut down their computers because of \" system difficulties . \"\nSentence 2: The New York Times asked employees at its headquarters to shut down their computers yesterday because of \" computing system difficulties . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The New York Times asked employees at its headquarters to shut down their computers yesterday because of \" computing system difficulties . \"\nSentence 2: Although it 's unclear whether Sobig was to blame , The New York Times also asked employees at its headquarters yesterday to shut down their computers because of \" system difficulties . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有什么方法可以减少电脑辐射？\nSentence 2: 什么方法可以减少电脑辐射呢", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么方法可以减少电脑辐射呢\nSentence 2: 有什么方法可以减少电脑辐射？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么品牌的染发剂最好\nSentence 2: 什么牌子的染发剂最好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么牌子的染发剂最好\nSentence 2: 什么品牌的染发剂最好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Female in a military uniform with her hair tied back.\nSentence 2: the female is wearing a ponytail", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: the female is wearing a ponytail\nSentence 2: Female in a military uniform with her hair tied back.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 此图出自哪里\nSentence 2: 此图出自何处？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 此图出自何处？\nSentence 2: 此图出自哪里", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该站于1903年7月1日在Stoneorlar至Glenties的多尼戈尔铁路线上开通。\nSentence 2: 该站于1903年7月1日在多尼戈尔铁路公司的线路上开通，从Glenties到Stranorlar。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该站于1903年7月1日在多尼戈尔铁路公司的线路上开通，从Glenties到Stranorlar。\nSentence 2: 该站于1903年7月1日在Stoneorlar至Glenties的多尼戈尔铁路线上开通。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppo x\nSentence 2: oppofx", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppofx\nSentence 2: oppo x", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人把一种混合物倒进一个碗里，然后把它混合在一起。\nSentence 2: 一个女人正在煎锅里煮鸡蛋。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人正在煎锅里煮鸡蛋。\nSentence 2: 一个女人把一种混合物倒进一个碗里，然后把它混合在一起。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman in red blouse is standing with small blond child in front of a small folding chalkboard.\nSentence 2: a woman stands with a child", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a woman stands with a child\nSentence 2: A woman in red blouse is standing with small blond child in front of a small folding chalkboard.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 找朋友借钱怎么开口说呢？\nSentence 2: 问朋友借钱好难，好难开口怎么办？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 问朋友借钱好难，好难开口怎么办？\nSentence 2: 找朋友借钱怎么开口说呢？", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 零点四五乘四\nSentence 2: 零点四五乘零点二", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 零点四五乘零点二\nSentence 2: 零点四五乘四", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A crowd wearing orange cheering for their team in a stadium.\nSentence 2: People wearing orange.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People wearing orange.\nSentence 2: A crowd wearing orange cheering for their team in a stadium.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is slicing bread\nSentence 2: A person is slicing a piece of bread", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is slicing a piece of bread\nSentence 2: A man is slicing bread", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two individuals use a photo kiosk\nSentence 2: There are two individuals", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are two individuals\nSentence 2: Two individuals use a photo kiosk", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: GOODGOOD\nSentence 2: Thatsgood", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Thatsgood\nSentence 2: GOODGOOD", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: HIBABY\nSentence 2: HIKING", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: HIKING\nSentence 2: HIBABY", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Mayor Joe T. Parker said late Thursday that the three workers were two men and a woman who were inside the building when the first blast occurred .\nSentence 2: The missing workers , two men and a woman , were inside the building when the first blast occurred , Mayor Joe T. Parker said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The missing workers , two men and a woman , were inside the building when the first blast occurred , Mayor Joe T. Parker said .\nSentence 2: Mayor Joe T. Parker said late Thursday that the three workers were two men and a woman who were inside the building when the first blast occurred .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of people are walking by a van.\nSentence 2: Some people outdoors.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some people outdoors.\nSentence 2: A group of people are walking by a van.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 37分钟后叫我\nSentence 2: 三十分钟计时开始", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 三十分钟计时开始\nSentence 2: 37分钟后叫我", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 印度的建筑倒塌；25人害怕被困。\nSentence 2: 印度建筑倒塌造成14人死亡，数十人被困。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 印度建筑倒塌造成14人死亡，数十人被困。\nSentence 2: 印度的建筑倒塌；25人害怕被困。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 发短信给妈\nSentence 2: 搜索信息", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 搜索信息\nSentence 2: 发短信给妈", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A crowd wearing orange cheering for their team in a stadium.\nSentence 2: Fans cheering on their team at the big game.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Fans cheering on their team at the big game.\nSentence 2: A crowd wearing orange cheering for their team in a stadium.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is applying makeup to her face\nSentence 2: There is no girl applying makeup to her face", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no girl applying makeup to her face\nSentence 2: A girl is applying makeup to her face", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Green tea consumption is associated with decreased risk of breast, pancreatic, colon, oesophageal, and lung cancers in humans.\nSentence 2: Tea protects from some diseases.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tea protects from some diseases.\nSentence 2: Green tea consumption is associated with decreased risk of breast, pancreatic, colon, oesophageal, and lung cancers in humans.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is quickly chopping some mushrooms with a knife\nSentence 2: A person is quickly chopping mushrooms with a knife", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is quickly chopping mushrooms with a knife\nSentence 2: A man is quickly chopping some mushrooms with a knife", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 二十年前\nSentence 2: 明月何时照我还", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明月何时照我还\nSentence 2: 二十年前", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The victims were not identified , although authorities said at least one of the bodies had been there about a year .\nSentence 2: The two victims buried in the yard were not identified , though authorities said at least one has been there about a year .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The two victims buried in the yard were not identified , though authorities said at least one has been there about a year .\nSentence 2: The victims were not identified , although authorities said at least one of the bodies had been there about a year .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: We feel so strongly about this issue that we are suspending sales and distribution of SCOLinux until these issues are resolved , \" Sontag said .\nSentence 2: We feel so strongly about this issue that we are suspending sales and distribution of SCO Linux until these issues are resolved . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: We feel so strongly about this issue that we are suspending sales and distribution of SCO Linux until these issues are resolved . \"\nSentence 2: We feel so strongly about this issue that we are suspending sales and distribution of SCOLinux until these issues are resolved , \" Sontag said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 小游戏app上找不到玩过的游戏\nSentence 2: 脚上有伤能拍上去几天就好的软件", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 脚上有伤能拍上去几天就好的软件\nSentence 2: 小游戏app上找不到玩过的游戏", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你是人吗\nSentence 2: 这是什么东西", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这是什么东西\nSentence 2: 你是人吗", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天你吃了晚饭吗\nSentence 2: 今天天气很晚饭吗", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气很晚饭吗\nSentence 2: 今天你吃了晚饭吗", "response": "1", "predicted": 1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is riding a horse in the desert\nSentence 2: A horse is being ridden by a guy", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A horse is being ridden by a guy\nSentence 2: A man is riding a horse in the desert", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A high school band getting ready to perform.\nSentence 2: A high school band prepares to play in front of a large crowd.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A high school band prepares to play in front of a large crowd.\nSentence 2: A high school band getting ready to perform.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor9splusota\nSentence 2: ￥lsquo", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: ￥lsquo\nSentence 2: oppor9splusota", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are eating\nSentence 2: A man and a woman are not eating", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are not eating\nSentence 2: A man and a woman are eating", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它沿着旧的Sohmer钢琴厂，在核桃街下，沿着主街。\nSentence 2: 它沿着老Sommer钢琴厂，在核桃街和主街沿线。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它沿着老Sommer钢琴厂，在核桃街和主街沿线。\nSentence 2: 它沿着旧的Sohmer钢琴厂，在核桃街下，沿着主街。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去学校\nSentence 2: 我明天去学孝", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我明天去学孝\nSentence 2: 我今天去学校", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜欢吃苹果\nSentence 2: 我喜焑吃苹果", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜焑吃苹果\nSentence 2: 我喜欢吃苹果", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The most common form of estrogen given to postmenopausal women is conjugated estrogen, which is made from the urine of pregnant mares, and the most popular brand of conjugated estrogen is Premarin.\nSentence 2: Estrogen is used by postmenopausal women.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Estrogen is used by postmenopausal women.\nSentence 2: The most common form of estrogen given to postmenopausal women is conjugated estrogen, which is made from the urine of pregnant mares, and the most popular brand of conjugated estrogen is Premarin.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no person boiling noodles\nSentence 2: A woman is boiling noodles in water", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is boiling noodles in water\nSentence 2: There is no person boiling noodles", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 谁有铠甲勇士之帝皇侠电影版\nSentence 2: 谁有铠甲勇士之帝皇侠电影，不要花絮。电影", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 谁有铠甲勇士之帝皇侠电影，不要花絮。电影\nSentence 2: 谁有铠甲勇士之帝皇侠电影版", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The government and private technology experts warned Wednesday that hackers plan to attack thousands of Web sites on Sunday in a loosely coordinated \" contest \" that could disrupt Internet traffic .\nSentence 2: THE US government and private technology experts have warned that hackers plan to attack thousands of websites on Sunday in a loosely co-ordinated \" contest \" that could disrupt Internet traffic .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: THE US government and private technology experts have warned that hackers plan to attack thousands of websites on Sunday in a loosely co-ordinated \" contest \" that could disrupt Internet traffic .\nSentence 2: The government and private technology experts warned Wednesday that hackers plan to attack thousands of Web sites on Sunday in a loosely coordinated \" contest \" that could disrupt Internet traffic .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man rides a black motorcycle while wearing a helmet and a van rides beside him in the distance.\nSentence 2: A motorcycle and a van ride next to each other.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A motorcycle and a van ride next to each other.\nSentence 2: A man rides a black motorcycle while wearing a helmet and a van rides beside him in the distance.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 七龙珠小说\nSentence 2: 月亮抱着太阳小说中文版", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 月亮抱着太阳小说中文版\nSentence 2: 七龙珠小说", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The flagship \"Atocha\" and eight other ships loaded with treasures went down, together with the crews of 550.\nSentence 2: People were killed in a collision between ships.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People were killed in a collision between ships.\nSentence 2: The flagship \"Atocha\" and eight other ships loaded with treasures went down, together with the crews of 550.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: When an earthquake rumbled off the coast of Hokkaido in Japan in July of 1993, the resulting tsunami hit just three to five minutes later, killing 202 people who were trying to flee for higher ground.\nSentence 2: An earthquake occurred on the coast of Hokkaido, Japan.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An earthquake occurred on the coast of Hokkaido, Japan.\nSentence 2: When an earthquake rumbled off the coast of Hokkaido in Japan in July of 1993, the resulting tsunami hit just three to five minutes later, killing 202 people who were trying to flee for higher ground.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 台湾新闻\nSentence 2: 美国选举新闻", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 美国选举新闻\nSentence 2: 台湾新闻", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Inception是一部2010年的科幻电影，由艾玛托马斯联合制作，由克里斯托弗诺兰编剧，联合制作和导演。\nSentence 2: Inception是一部2010年的科幻电影，由Emma Thomas联合制作，由Christopher Nolan编写，共同制作并上演。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Inception是一部2010年的科幻电影，由Emma Thomas联合制作，由Christopher Nolan编写，共同制作并上演。\nSentence 2: Inception是一部2010年的科幻电影，由艾玛托马斯联合制作，由克里斯托弗诺兰编剧，联合制作和导演。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在2015年的地区选举中，在托西被联邦党排除在外，吉多尼被选入贝卢诺省的威尼托区域委员会。\nSentence 2: 在2015年的地区选举中，在托西被联邦政党搁置后，吉多尼当选为贝卢诺省的威尼托区域委员会成员。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在2015年的地区选举中，在托西被联邦政党搁置后，吉多尼当选为贝卢诺省的威尼托区域委员会成员。\nSentence 2: 在2015年的地区选举中，在托西被联邦党排除在外，吉多尼被选入贝卢诺省的威尼托区域委员会。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气不错\nSentence 2: 今天天气不做", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气不做\nSentence 2: 今天天气不错", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Specifically, the October 4, 1994 earthquake (as the October 12, 1963 earthquake) ocurred at the Pacific side boundary of the Kuril arc (the South Kuril Islands) of the Okhotsk plate.\nSentence 2: An earthquake occurred in 1994.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An earthquake occurred in 1994.\nSentence 2: Specifically, the October 4, 1994 earthquake (as the October 12, 1963 earthquake) ocurred at the Pacific side boundary of the Kuril arc (the South Kuril Islands) of the Okhotsk plate.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 笑话我的\nSentence 2: 给我讲个笑话好不好", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 给我讲个笑话好不好\nSentence 2: 笑话我的", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我们来玩成语接龙\nSentence 2: 成语接龙大全", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 成语接龙大全\nSentence 2: 我们来玩成语接龙", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我这配置能玩洛奇英雄传吗？\nSentence 2: 我这配置可以玩洛奇英雄传吗", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我这配置可以玩洛奇英雄传吗\nSentence 2: 我这配置能玩洛奇英雄传吗？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is not walking the dog\nSentence 2: The woman is walking the dog", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is walking the dog\nSentence 2: The woman is not walking the dog", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃兔肉吗\nSentence 2: 怀孕能吃香菜吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃香菜吗\nSentence 2: 怀孕能吃兔肉吗", "response": "1", "predicted": 1, "true_label": 1}

eval_judge/gpu_results/results_gpu_4.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 吃柿子有什么好处\nSentence 2: 怀孕能吃小龙虾吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃小龙虾吗\nSentence 2: 吃柿子有什么好处", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 又向自己的脸颊和鼻子使劲打了五六拳，打得血肉模糊。\nSentence 2: 于是他两手去抓自己的胡子，把胡子揪掉了一半，然后又向自己的面颊和鼻子一连打了五六拳，打得自己满脸是血。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 于是他两手去抓自己的胡子，把胡子揪掉了一半，然后又向自己的面颊和鼻子一连打了五六拳，打得自己满脸是血。\nSentence 2: 又向自己的脸颊和鼻子使劲打了五六拳，打得血肉模糊。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 德国一季度实现增长，躲过经济衰退\nSentence 2: 多亏了德国，欧元区避免了第一季度的衰退。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 多亏了德国，欧元区避免了第一季度的衰退。\nSentence 2: 德国一季度实现增长，躲过经济衰退", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样炒青菜才好吃啊？\nSentence 2: 青菜怎么炒才好吃?", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 青菜怎么炒才好吃?\nSentence 2: 怎样炒青菜才好吃啊？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: When it was revealed that under her government, irregular special forces, possibly including the Hezbollah, were illegally financed to fight the PKK, Ciller reacted by saying she had done no wrong and would do the same again.\nSentence 2: A French political party was illegally financed.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A French political party was illegally financed.\nSentence 2: When it was revealed that under her government, irregular special forces, possibly including the Hezbollah, were illegally financed to fight the PKK, Ciller reacted by saying she had done no wrong and would do the same again.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 圆脸小眼睛女孩适合什么发型\nSentence 2: 圆脸小眼睛不高适合什么发型和刘海。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 圆脸小眼睛不高适合什么发型和刘海。\nSentence 2: 圆脸小眼睛女孩适合什么发型", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 英国马恩岛航空公司的历史可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。\nSentence 2: 马恩岛航空公司可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 马恩岛航空公司可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。\nSentence 2: 英国马恩岛航空公司的历史可以追溯到1991年3月，当时马恩岛航空公司创建了欧洲地区航空公司，以便在英国境内扩展和飞行航线。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 美女主播写真集\nSentence 2: 赛文奥特曼的照片", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 赛文奥特曼的照片\nSentence 2: 美女主播写真集", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 娜塔莎再也不跟索尼娅说话，躲着她。\nSentence 2: 娜塔莎不再跟索尼娅讲话了，并且老躲着她。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 娜塔莎不再跟索尼娅讲话了，并且老躲着她。\nSentence 2: 娜塔莎再也不跟索尼娅说话，躲着她。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打开什么值得买里签到好不好\nSentence 2: 那你在干嘛", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 那你在干嘛\nSentence 2: 打开什么值得买里签到好不好", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 阿富汗首都自杀式炸弹袭击造成6人死亡\nSentence 2: 自杀式袭击者在阿富汗清真寺炸死37人", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 自杀式袭击者在阿富汗清真寺炸死37人\nSentence 2: 阿富汗首都自杀式炸弹袭击造成6人死亡", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 刻薄的反义词是什么\nSentence 2: 刻薄的反义词是什么？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 刻薄的反义词是什么？\nSentence 2: 刻薄的反义词是什么", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 送前男友什么生日礼��好\nSentence 2: 送什么生日礼物给前男友好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 送什么生日礼物给前男友好\nSentence 2: 送前男友什么生日礼物好", "response": "4", "predicted": 4, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The former president also gave numerous speeches in 2002 without compensation , said his spokesman Jim Kennedy .\nSentence 2: His spokesman Jim Kennedy said the former president also gave more than 70 speeches in 2002 without compensation .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: His spokesman Jim Kennedy said the former president also gave more than 70 speeches in 2002 without compensation .\nSentence 2: The former president also gave numerous speeches in 2002 without compensation , said his spokesman Jim Kennedy .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: When replacing a single tooth, an implant has an advantage over fixed bridgework in that healthy teeth adjacent to the missing tooth do not have to be prepped (drilled down).\nSentence 2: Tooth implants have advantages.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tooth implants have advantages.\nSentence 2: When replacing a single tooth, an implant has an advantage over fixed bridgework in that healthy teeth adjacent to the missing tooth do not have to be prepped (drilled down).", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: AMSORRY\nSentence 2: IMSORRY", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: IMSORRY\nSentence 2: AMSORRY", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 界上最小的国家\nSentence 2: 界上有鬼吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 界上有鬼吗\nSentence 2: 界上最小的国家", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is singing and playing a guitar\nSentence 2: A man is playing a guitar", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is playing a guitar\nSentence 2: A man is singing and playing a guitar", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Female gymnasts warm up before a competition.\nSentence 2: Gymnasts get ready for a competition.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Gymnasts get ready for a competition.\nSentence 2: Female gymnasts warm up before a competition.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天天酷跑百变萝莉和超能少年哪个好\nSentence 2: 天天酷跑超能少年和百变萝莉哪个更好一点", "response": "4", "predicted": 4, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天天酷跑超能少年和百变萝莉哪个更好一点\nSentence 2: 天天酷跑百变萝莉和超能少年哪个好", "response": "4", "predicted": 4, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her.\nSentence 2: The young girl is wearing a red shirt.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The young girl is wearing a red shirt.\nSentence 2: A young girl wearing a red shirt and cap smiling and holding a small toy is standing in front of a group of children playing behind her.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Gateway will release new Profile 4 systems with the new Intel technology on Wednesday .\nSentence 2: Gateway 's all-in-one PC , the Profile 4 , also now features the new Intel technology .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Gateway 's all-in-one PC , the Profile 4 , also now features the new Intel technology .\nSentence 2: Gateway will release new Profile 4 systems with the new Intel technology on Wednesday .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样举报拼多多客服\nSentence 2: 软件太多费流量怎么办", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 软件太多费流量怎么办\nSentence 2: 怎样举报拼多多客服", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Bogdanowice是波兰的一个村庄，位于Głubczyce县（CommuneGłubczyce）的Opole Voivodeship。\nSentence 2: Bogdanowice是一个位于波兰Głubczyce县的村庄，位于Opole Voivodeship（GminaGłubczyce）。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Bogdanowice是一个位于波兰Głubczyce县的村庄，位于Opole Voivodeship（GminaGłubczyce）。\nSentence 2: Bogdanowice是波兰的一个村庄，位于Głubczyce县（CommuneGłubczyce）的Opole Voivodeship。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 九千公顷等于多少平方千米\nSentence 2: 分钟等于多少秒", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 分钟等于多少秒\nSentence 2: 九千公顷等于多少平方千米", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 巴基斯坦汽车炸弹炸死17人\nSentence 2: 巴基斯坦自杀式袭击者炸死10人", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 巴基斯坦自杀式袭击者炸死10人\nSentence 2: 巴基斯坦汽车炸弹炸死17人", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three men are dancing\nSentence 2: The men are not dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The men are not dancing\nSentence 2: Three men are dancing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is stirring meat in a bowl\nSentence 2: There is no lady mixing a meat mixture in a bowl", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no lady mixing a meat mixture in a bowl\nSentence 2: The woman is stirring meat in a bowl", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Zotov said that Russia would agree to the complete withdrawal of its troops from Latvia by August 31, 1994 if Latvia allowed Russia to use the Skrunda radar for the next four years and dismantle it within the following 18 months.\nSentence 2: Russian troops withdrew from Latvia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Russian troops withdrew from Latvia.\nSentence 2: Zotov said that Russia would agree to the complete withdrawal of its troops from Latvia by August 31, 1994 if Latvia allowed Russia to use the Skrunda radar for the next four years and dismantle it within the following 18 months.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPO r11s plus\nSentence 2: oppor11plus", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor11plus\nSentence 2: OPPO r11s plus", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Man and woman rowing on surfboards third rower in distance.\nSentence 2: Couples rowing on surfboards.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Couples rowing on surfboards.\nSentence 2: Man and woman rowing on surfboards third rower in distance.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is fixing a silencer to a gun\nSentence 2: A man is applying a silencer to a gun", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is applying a silencer to a gun\nSentence 2: A man is fixing a silencer to a gun", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 教堂边门口，他立住脚，发现我上气不接下气。\nSentence 2: 在教堂墓地的边门口，他停了下来；他发现我已经完全上气不接下气了。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在教堂墓地的边门口，他停了下来；他发现我已经完全上气不接下气了。\nSentence 2: 教堂边门口，他立住脚，发现我上气不接下气。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no girl playing the piano\nSentence 2: A girl is playing the piano", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A girl is playing the piano\nSentence 2: There is no girl playing the piano", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A kitten is playing with a blue rope toy\nSentence 2: The kitten is not playing with a toy", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The kitten is not playing with a toy\nSentence 2: A kitten is playing with a blue rope toy", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two men are fighting in a cow pasture\nSentence 2: Nobody is fighting in a cow pasture", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nobody is fighting in a cow pasture\nSentence 2: Two men are fighting in a cow pasture", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is boiling noodles\nSentence 2: There is no person boiling noodles", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no person boiling noodles\nSentence 2: A person is boiling noodles", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A roofer in a gray sweatshirt and orange hat walks on a unfinished roof at a lake-side home.\nSentence 2: The roofer is outside working.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The roofer is outside working.\nSentence 2: A roofer in a gray sweatshirt and orange hat walks on a unfinished roof at a lake-side home.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Italy's Alberto Tomba pulled out of the 2nd race.\nSentence 2: Alberto Tomba won a ski race.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Alberto Tomba won a ski race.\nSentence 2: Italy's Alberto Tomba pulled out of the 2nd race.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOZMXP\nSentence 2: oppo x", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppo x\nSentence 2: OPPOZMXP", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is cutting a box\nSentence 2: There is no man cutting a box", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man cutting a box\nSentence 2: A man is cutting a box", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: areyouok\nSentence 2: whatdoyoudo", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: whatdoyoudo\nSentence 2: areyouok", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 去哪个网站可以免费下载歌曲\nSentence 2: 哪个网站可以免费下载完整版歌曲", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 哪个网站可以免费下载完整版歌曲\nSentence 2: 去哪个网站可以免费下载歌曲", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公园散步了。\nSentence 2: 我今天去公元散步了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公元散步了。\nSentence 2: 我今天去公园散步了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一只湿漉漉的金发狗在岸边举着一根棍子。\nSentence 2: 这只棕色湿漉漉的狗嘴里叼着一根棍子，正在水旁的沙子里奔跑。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这只棕色湿漉漉的狗嘴里叼着一根棍子，正在水旁的沙子里奔跑。\nSentence 2: 一只湿漉漉的金发狗在岸边举着一根棍子。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three women are sitting on a green bench looking out at the coast.\nSentence 2: The women are outdoors.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The women are outdoors.\nSentence 2: Three women are sitting on a green bench looking out at the coast.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 几点了呀\nSentence 2: 感恩节是几月几号", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 感恩节是几月几号\nSentence 2: 几点了呀", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A mountainous photo is complete with a blue sky.\nSentence 2: Someone has taken a photo.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone has taken a photo.\nSentence 2: A mountainous photo is complete with a blue sky.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Elderly woman in blue apron balances a basket on her head on a sidewalk while talking to a woman dressed in black.\nSentence 2: Elderly woman is balancing something on her head while having a conversation.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Elderly woman is balancing something on her head while having a conversation.\nSentence 2: Elderly woman in blue apron balances a basket on her head on a sidewalk while talking to a woman dressed in black.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 十一月四日风雨大作\nSentence 2: 长风破浪会有时下一句", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 长风破浪会有时下一句\nSentence 2: 十一月四日风雨大作", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 练瑜伽对女人有什么好处？\nSentence 2: 练瑜伽对女人有什么好处?", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 练瑜伽对女人有什么好处?\nSentence 2: 练瑜伽对女人有什么好处？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 尽管存在区域性财务影响，但似乎还有压倒性的社会模式继续这一趋势。\nSentence 2: 尽管存在区域性财务影响，但似乎仍有压倒性的社会模式使这一趋势长期存在。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 尽管存在区域性财务影响，但似乎仍有压倒性的社会模式使这一趋势长期存在。\nSentence 2: 尽管存在区域性财务影响，但似乎还有压倒性的社会模式继续这一趋势。", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A family walking with a soldier.\nSentence 2: A group of people strolling.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of people strolling.\nSentence 2: A family walking with a soldier.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天是个晴天吗\nSentence 2: 今天是个晴天马", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天是个晴天马\nSentence 2: 今天是个晴天吗", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党议员和前民主党领袖组成。\nSentence 2: PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党长和前党派领导人组成。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党长和前党派领导人组成。\nSentence 2: PLEO代表通常由民主党全国委员会成员，民主党国会议员，民主党议员和前民主党领袖组成。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第 46 火箭师于 1941 年 4 月 29 日创立，当时第 188 步兵师已于考纳斯完成组建。\nSentence 2: 第 46 火箭师的历史始于 1941 年 4 月 29 日，当时第 188 步枪师在考纳斯组建完成。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第 46 火箭师的历史始于 1941 年 4 月 29 日，当时第 188 步枪师在考纳斯组建完成。\nSentence 2: 第 46 火箭师于 1941 年 4 月 29 日创立，当时第 188 步兵师已于考纳斯完成组建。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of young adults are standing on a street in front of a brick wall with a tall wooden gate.\nSentence 2: A group of friends poses for a picture in front of a brick wall.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of friends poses for a picture in front of a brick wall.\nSentence 2: A group of young adults are standing on a street in front of a brick wall with a tall wooden gate.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The rhino is grazing on the grass\nSentence 2: The rhino is being fed on the grass", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The rhino is being fed on the grass\nSentence 2: The rhino is grazing on the grass", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a suit driving a horse-drawn buggy down a stone street.\nSentence 2: The man is driving a buggy.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is driving a buggy.\nSentence 2: A man in a suit driving a horse-drawn buggy down a stone street.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 向Borsalino致敬的是Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑。\nSentence 2: 由Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑都是对Borsalino的致敬。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 由Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑都是对Borsalino的致敬。\nSentence 2: 向Borsalino致敬的是Philippe Starck为Flos设计的Chapeau Lamp（2014）和由Moritz Waldemeyer设计的The Hatband（2016）雕塑。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 三个孩子在雪中玩耍。\nSentence 2: 三个孩子在干草里玩耍", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 三个孩子在干草里玩耍\nSentence 2: 三个孩子在雪中玩耍。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy swims shirtless with a black and red bathing suit while a seagull sits on the sand.\nSentence 2: A boy swims.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy swims.\nSentence 2: A boy swims shirtless with a black and red bathing suit while a seagull sits on the sand.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你妹妹是一条龙\nSentence 2: 我是一条龙", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我是一条龙\nSentence 2: 你妹妹是一条龙", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The head of the US Democratic Party has lambasted Mankiw, and suggested that he be outsourced to India.\nSentence 2: Jobs were outsourced to India.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Jobs were outsourced to India.\nSentence 2: The head of the US Democratic Party has lambasted Mankiw, and suggested that he be outsourced to India.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 哥哥姐姐们，拜托了\nSentence 2: 哥哥姐姐们，拜托了，", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 哥哥姐姐们，拜托了，\nSentence 2: 哥哥姐姐们，拜托了", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么烧猪蹄好吃呢？\nSentence 2: 猪蹄炖什么才好吃", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 猪蹄炖什么才好吃\nSentence 2: 什么烧猪蹄好吃呢？", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: drop\nSentence 2: probebly", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: probebly\nSentence 2: drop", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: WHATAREYOUGOINGTODO\nSentence 2: Whereareyougoing", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whereareyougoing\nSentence 2: WHATAREYOUGOINGTODO", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 梯形的高怎么求\nSentence 2: 第二笔是什么", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 第二笔是什么\nSentence 2: 梯形的高怎么求", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The government thereby perpetuates our continued dependence on fuels that are environmentally and economically unsustainable.\nSentence 2: The government took measures to protect the environment.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The government took measures to protect the environment.\nSentence 2: The government thereby perpetuates our continued dependence on fuels that are environmentally and economically unsustainable.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天我去上课\nSentence 2: 明天我去上个", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天我去上个\nSentence 2: 明天我去上课", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这是一个测试文本\nSentence 2: 这是一亇测试文本", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这是一亇测试文本\nSentence 2: 这是一个测试文本", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man wearing a hat and boots is digging for something in the snow.\nSentence 2: A man is trying to get to something by digging.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is trying to get to something by digging.\nSentence 2: A man wearing a hat and boots is digging for something in the snow.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A team of men, three in red shirts and two others in baseball jerseys, re-surface a baseball diamond with specialized tools.\nSentence 2: A grounds crew maintain a park.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A grounds crew maintain a park.\nSentence 2: A team of men, three in red shirts and two others in baseball jerseys, re-surface a baseball diamond with specialized tools.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它由Sakamoto编写，由瑞典歌手Frida Sundemo创作。\nSentence 2: 它由Sakamoto创作，由瑞典歌手Frida Sundemo编写。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它由Sakamoto创作，由瑞典歌手Frida Sundemo编写。\nSentence 2: 它由Sakamoto编写，由瑞典歌手Frida Sundemo创作。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: llll\nSentence 2: lllll", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: lllll\nSentence 2: llll", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tony-award winning dancer and actor Gregory Hines died of cancer Saturday in Los Angeles .\nSentence 2: Hines died yesterday in Los Angeles of cancer , publicist Allen Eichorn said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Hines died yesterday in Los Angeles of cancer , publicist Allen Eichorn said .\nSentence 2: Tony-award winning dancer and actor Gregory Hines died of cancer Saturday in Los Angeles .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样写好写快钢笔字\nSentence 2: 怎样写好钢笔字。", "response": "合并后的句子：怎样写好钢笔字。\n\n1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样写好钢笔字。\nSentence 2: 怎样写好写快钢笔字", "response": "-1", "predicted": -1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: WiFi6\nSentence 2: wifi", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: wifi\nSentence 2: WiFi6", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天天气tianqi怎么样？\nSentence 2: 明天的天气将会很hao。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天的天气将会很hao。\nSentence 2: 明天天气tianqi怎么样？", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is sharpening a knife\nSentence 2: A man is sharpening a weapon", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is sharpening a weapon\nSentence 2: A man is sharpening a knife", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两只动物坐在岩石上。\nSentence 2: 两只老鹰坐在一根长长的树枝上。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两只老鹰坐在一根长长的树枝上。\nSentence 2: 两只动物坐在岩石上。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head.\nSentence 2: A couple are having a photoshoot.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A couple are having a photoshoot.\nSentence 2: A man and woman in formal attire are standing in a high grass meadow, posing for the photographer, and she has two of her fingers up on one hand behind his head.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Generally the coffee or tea drinker feels worse in just a few short minutes after finishing the drink.\nSentence 2: Tea protects from illness.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Tea protects from illness.\nSentence 2: Generally the coffee or tea drinker feels worse in just a few short minutes after finishing the drink.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a young boy using a field microscope to identify a field specimen during a field trip.\nSentence 2: The boy is looking through a microscope.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The boy is looking through a microscope.\nSentence 2: a young boy using a field microscope to identify a field specimen during a field trip.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: One of the main treatments used to combat the symptoms of mental illness is medication in various forms.\nSentence 2: Drugs are used to combat schizophrenia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Drugs are used to combat schizophrenia.\nSentence 2: One of the main treatments used to combat the symptoms of mental illness is medication in various forms.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 北京现代怎么样？\nSentence 2: 北京现代怎么样", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 北京现代怎么样\nSentence 2: 北京现代怎么样？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机充电后自动关机开机后就一直打不开\nSentence 2: 手机没电关机后充电一直震动无法开机", "response": "1", "predicted": 1, "true_label": 1}

eval_judge/gpu_results/results_gpu_5.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机没电关机后充电一直震动无法开机\nSentence 2: 手机充电后自动关机开机后就一直打不开", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Mansfield attempted to list every vessel that ever sailed the Lakes before 1899, as well as every shipwreck.\nSentence 2: Shipwreck salvaging was attempted.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Shipwreck salvaging was attempted.\nSentence 2: Mansfield attempted to list every vessel that ever sailed the Lakes before 1899, as well as every shipwreck.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A shirtless man is painting over a graffiti filled wall with a paint roller.\nSentence 2: A man is painting a wall.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is painting a wall.\nSentence 2: A shirtless man is painting over a graffiti filled wall with a paint roller.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 叙利亚要求保证，反政府武装称和平计划注定要失败。\nSentence 2: 叙利亚要求提供撤军的书面保证", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 叙利亚要求提供撤军的书面保证\nSentence 2: 叙利亚要求保证，反政府武装称和平计划注定要失败。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is riding a horse in a sandy land\nSentence 2: A guy is riding a horse", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A guy is riding a horse\nSentence 2: A man is riding a horse in a sandy land", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 离这儿不远有一座寺庙，表弟说，里面住着一位修士，据说，此人当过兵，口碑不错，而且还非常谦和善良。\nSentence 2: 那位表亲说:离这儿不远，住着个隐居的修士，据说当过兵，大家认为他是个很有见识的好基督徒，对人也很诚实。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 那位表亲说:离这儿不远，住着个隐居的修士，据说当过兵，大家认为他是个很有见识的好基督徒，对人也很诚实。\nSentence 2: 离这儿不远有一座寺庙，表弟说，里面住着一位修士，据说，此人当过兵，口碑不错，而且还非常谦和善良。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在线宠物医生咨询\nSentence 2: 在线宠物医生咨询急！", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在线宠物医生咨询急！\nSentence 2: 在线宠物医生咨询", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好啊\nSentence 2: 今天天啊好真", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天啊好真\nSentence 2: 今天天气真好啊", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样关闭防止游戏提示\nSentence 2: 总提示软件有危险怎么关闭", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 总提示软件有危险怎么关闭\nSentence 2: 怎样关闭防止游戏提示", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 穿绿色运动衫的男孩在足球队担任守门员。\nSentence 2: 穿蓝衣服的球队正在和穿白色衣服的球队比赛。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 穿蓝衣服的球队正在和穿白色衣服的球队比赛。\nSentence 2: 穿绿色运动衫的男孩在足球队担任守门员。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: And his justification for the bombing was that while it caused short-term material damage , it was for the long-term moral good of Bali .\nSentence 2: And he justified bombing Bali by saying that while it had caused material devastation , it was for the island 's long-term moral good .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: And he justified bombing Bali by saying that while it had caused material devastation , it was for the island 's long-term moral good .\nSentence 2: And his justification for the bombing was that while it caused short-term material damage , it was for the long-term moral good of Bali .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有几个可能的原因。\nSentence 2: 画皱纹有几个原因。", "response": "1", "predicted": 1, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 画皱纹有几个原因。\nSentence 2: 有几个可能的原因。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两个女人站在旅游车前。\nSentence 2: 两个人在飞机前说话。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两个人在飞机前说话。\nSentence 2: 两个女人站在旅游车前。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机录屏黑屏是什么原因\nSentence 2: 手机黑屏关机是什么原因", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机黑屏关机是什么原因\nSentence 2: 手机录屏黑屏是什么原因", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 人们试图将修道院的纳扎尔人解释为对禁欲训练实践上瘾的圣经命令的先驱。\nSentence 2: 已经尝试解释修道士纳扎尔人作为圣经命令的先驱，他们沉迷于禁欲训练的实践。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 已经尝试解释修道士纳扎尔人作为圣经命令的先驱，他们沉迷于禁欲训练的实践。\nSentence 2: 人们试图将修道院的纳扎尔人解释为对禁欲训练实践上瘾的圣经命令的先驱。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Aksuat Dendropark（俄语：Akshut Dendroparks）是一座受保护的纪念碑（Ulyanovsk Oblast的自然区域）。\nSentence 2: Aksuat Dendropark（俄语：Akshout Dendroparks）是一个天然纪念物（乌里扬诺夫斯克州保护区）", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Aksuat Dendropark（俄语：Akshout Dendroparks）是一个天然纪念物（乌里扬诺夫斯克州保护区）\nSentence 2: Aksuat Dendropark（俄语：Akshut Dendroparks）是一座受保护的纪念碑（Ulyanovsk Oblast的自然区域）。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Revenues for \" The Hulk \" came in well below those of last month 's Marvel Comics adaptation , \" X2 : X-Men United , \" which grossed $ 85.6 million in its opening weekend .\nSentence 2: The Hulk trailed last month 's Marvel Comics adaptation , X2 : X-Men United , which grossed $ 85.6-million in its opening weekend .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The Hulk trailed last month 's Marvel Comics adaptation , X2 : X-Men United , which grossed $ 85.6-million in its opening weekend .\nSentence 2: Revenues for \" The Hulk \" came in well below those of last month 's Marvel Comics adaptation , \" X2 : X-Men United , \" which grossed $ 85.6 million in its opening weekend .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" These foods have an almost identical effect on lowering cholesterol as the original cholesterol-lowering drugs . \"\nSentence 2: We have now proven that these foods have an almost identical effect on lowering cholesterol as the original cholesterol-reducing drugs .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: We have now proven that these foods have an almost identical effect on lowering cholesterol as the original cholesterol-reducing drugs .\nSentence 2: \" These foods have an almost identical effect on lowering cholesterol as the original cholesterol-lowering drugs . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man on a motorcycle is riding on one wheel\nSentence 2: A person is performing acrobatics on a motorcycle", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is performing acrobatics on a motorcycle\nSentence 2: A man on a motorcycle is riding on one wheel", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 电影中的岛屿位于埃尔斯米尔岛以北，而不是帕特里克王子岛。\nSentence 2: 该岛屿位于埃尔斯米尔岛以北的电影中，而不是帕特里克王子岛。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该岛屿位于埃尔斯米尔岛以北的电影中，而不是帕特里克王子岛。\nSentence 2: 电影中的岛屿位于埃尔斯米尔岛以北，而不是帕特里克王子岛。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Oppoa7\nSentence 2: opoopa73", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: opoopa73\nSentence 2: Oppoa7", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of men is playing soccer on the beach\nSentence 2: A group of men is playing with a ball on the beach", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of men is playing with a ball on the beach\nSentence 2: A group of men is playing soccer on the beach", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打麻将怎么赢钱\nSentence 2: 打麻将怎么才能赢钱？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打麻将怎么才能赢钱？\nSentence 2: 打麻将怎么赢钱", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 关于夏天的谚语\nSentence 2: 关于读书的名人名言", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 关于读书的名人名言\nSentence 2: 关于夏天的谚语", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天氣真好，我們去公園散步吧\nSentence 2: 天氣真號，我們去公園散步吧", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天氣真號，我們去公園散步吧\nSentence 2: 今天天氣真好，我們去公園散步吧", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: WHATSWRONG\nSentence 2: Whatswrongwithyou", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whatswrongwithyou\nSentence 2: WHATSWRONG", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一年好景君须记，下一句是什么\nSentence 2: 一年好景君须记的下一句是什么？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一年好景君须记的下一句是什么？\nSentence 2: 一年好景君须记，下一句是什么", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman slicing an onion\nSentence 2: A woman is slicing an onion", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is slicing an onion\nSentence 2: There is no woman slicing an onion", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你好小哦\nSentence 2: 你的小仓鼠呢", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你的小仓鼠呢\nSentence 2: 你好小哦", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman looks at a plate filled with steam.\nSentence 2: The woman is in a kitchen.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is in a kitchen.\nSentence 2: A woman looks at a plate filled with steam.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 外祖母满心感激地说。\nSentence 2: 外祖母感激地说。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 外祖母感激地说。\nSentence 2: 外祖母满心感激地说。", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The nightmare of human disaster in Rwanda in 1994, after UN Peacekeeping Forces withdrew, shocked us into questioning the whole concept of peacekeeping.\nSentence 2: UN peacekeeping forces withdrew from Bosnia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: UN peacekeeping forces withdrew from Bosnia.\nSentence 2: The nightmare of human disaster in Rwanda in 1994, after UN Peacekeeping Forces withdrew, shocked us into questioning the whole concept of peacekeeping.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她们两个有时单人独唱，有时二人合唱，我们很可以算得开了一个小小的音乐会。\nSentence 2: 她俩一起唱，还进行二声部合唱，这一来我们就举行了一个很圆满的小型音乐会。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她俩一起唱，还进行二声部合唱，这一来我们就举行了一个很圆满的小型音乐会。\nSentence 2: 她们两个有时单人独唱，有时二人合唱，我们很可以算得开了一个小小的音乐会。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Five people standing on a stage, getting ready to perform, who are unrecognizable due to the lack of lighting.\nSentence 2: In a dim room, several actors who can barely see each other, are getting ready to act.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: In a dim room, several actors who can barely see each other, are getting ready to act.\nSentence 2: Five people standing on a stage, getting ready to perform, who are unrecognizable due to the lack of lighting.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a soccer player with black and red striped uniform gets ready to kick the soccer ball\nSentence 2: a soccer player gets ready to kick the ball after a penalty.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a soccer player gets ready to kick the ball after a penalty.\nSentence 2: a soccer player with black and red striped uniform gets ready to kick the soccer ball", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它以Robert Suhosky的小说为基础，并由James Hardiman改编成剧本。\nSentence 2: 它是基于罗伯特·苏霍斯基的一部小说改编成詹姆斯·哈迪曼的剧本。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 它是基于罗伯特·苏霍斯基的一部小说改编成詹姆斯·哈迪曼的剧本。\nSentence 2: 它以Robert Suhosky的小说为基础，并由James Hardiman改编成剧本。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的手机在阳光下屏一点看不见\nSentence 2: 我的手机屏幕碎了打开后也点不开", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的手机屏幕碎了打开后也点不开\nSentence 2: 我的手机在阳光下屏一点看不见", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 最后，在1601年，出版了委托散文故事（“Bataviae Hollandiaeque Annales”）。\nSentence 2: 最后，出版的散文历史（“Bataviae Hollandiaeque Annales”）于1601年投入使用。", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 最��，出版的散文历史（“Bataviae Hollandiaeque Annales”）于1601年投入使用。\nSentence 2: 最后，在1601年，出版了委托散文故事（“Bataviae Hollandiaeque Annales”）。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怀孕能吃毛笋吗\nSentence 2: 泥鳅吃什么", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 泥鳅吃什么\nSentence 2: 怀孕能吃毛笋吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Violin soloists take the stage during the orchestra's opening show at the theater.\nSentence 2: People are playing violin on stage.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People are playing violin on stage.\nSentence 2: Violin soloists take the stage during the orchestra's opening show at the theater.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: people standing at a beach with Cameras.\nSentence 2: A group of people standing at a beach filled with cameras.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of people standing at a beach filled with cameras.\nSentence 2: people standing at a beach with Cameras.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Iraq 's nuclear program had been dismantled and there was no convincing evidence it was being revived , the report said .\nSentence 2: Iraq 's nuclear program had been dismantled , and there \" was no convincing evidence of its reconstitution . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Iraq 's nuclear program had been dismantled , and there \" was no convincing evidence of its reconstitution . \"\nSentence 2: Iraq 's nuclear program had been dismantled and there was no convincing evidence it was being revived , the report said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 四只黑狗跑过草地。\nSentence 2: 一只狗跑过草地。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一只狗跑过草地。\nSentence 2: 四只黑狗跑过草地。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A cat is sliding upside down under a couch\nSentence 2: An animal is crawling under a piece of furniture", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An animal is crawling under a piece of furniture\nSentence 2: A cat is sliding upside down under a couch", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 越南探测失踪飞机信号\nSentence 2: 中国扩大失踪飞机搜寻范围", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国扩大失踪飞机搜寻范围\nSentence 2: 越南探测失踪飞机信号", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在数学天文学领域，他的声誉归功于天文球体的提出，以及他对理解行星运动的早期贡献。\nSentence 2: 他在数学天文学方面享有盛誉是因为他引入了天文地球仪，并对理解行星运动作出了早期贡献。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他在数学天文学方面享有盛誉是因为他引入了天文地球仪，并对理解行星运动作出了早期贡献。\nSentence 2: 在数学天文学领域，他的声誉归功于天文球体的提出，以及他对理解行星运动的早期贡献。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这个美女叫什么名字\nSentence 2: 请问这个美女叫什么名字！", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请问这个美女叫什么名字！\nSentence 2: 请问这个美女叫什么名字", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of children divided into a maroon team and a blue team playing a soccer game.\nSentence 2: Kids are playing soccer.", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Kids are playing soccer.\nSentence 2: A group of children divided into a maroon team and a blue team playing a soccer game.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天去公园玩了。\nSentence 2: 我公元去玩昨天了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我公元去玩昨天了。\nSentence 2: 我昨天去公园玩了。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜欢打篮球。\nSentence 2: 我喜欢打蓝球。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜欢打蓝球。\nSentence 2: 我喜欢打篮球。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 20世纪90年代，在亚历山德里亚，比耶拉，伊夫雷亚，蒙多维和韦尔切利开设了新的教学场所。\nSentence 2: 在20世纪90年代，在Ivrea，Mondovì，Biella，Alessandria和Vercelli开设了新的教学园。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在20世纪90年代，在Ivrea，Mondovì，Biella，Alessandria和Vercelli开设了新的教学园。\nSentence 2: 20世纪90年代，在亚历山德里亚，比耶拉，伊夫雷亚，蒙多维和韦尔切利开设了新的教学场所。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Jack Ferry , company spokesman , said a search is ongoing but would not comment on the status .\nSentence 2: Company spokesman Jack Ferry said a search is ongoing but declined comment on its status .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Company spokesman Jack Ferry said a search is ongoing but declined comment on its status .\nSentence 2: Jack Ferry , company spokesman , said a search is ongoing but would not comment on the status .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is not riding a mechanical bull\nSentence 2: The man is riding a mechanical bull", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is riding a mechanical bull\nSentence 2: The man is not riding a mechanical bull", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The Government's new policies give more weight to reducing the adverse impact of trunk roads on the environment, including reducing greenhouse gas emissions and pollutants from trunk road traffic.\nSentence 2: The government took measures to reduce pollution.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The government took measures to reduce pollution.\nSentence 2: The Government's new policies give more weight to reducing the adverse impact of trunk roads on the environment, including reducing greenhouse gas emissions and pollutants from trunk road traffic.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去商店买了许多食物。\nSentence 2: 我今天去商店买了许多试物。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去商店买了许多试物。\nSentence 2: 我今天去商店买了许多食物。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我想去公园散步。\nSentence 2: 今天天气真好，我想去公元散步。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我想去公元散步。\nSentence 2: 今天天气真好，我想去公园散步。", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Australia , Chile , Colombia , El Salvador , Honduras , Mexico , New Zealand , Peru and Uruguay will also support the challenge .\nSentence 2: Nine other countries , including Australia , Chile , Colombia , El Salvador and Mexico , are supporting the case .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nine other countries , including Australia , Chile , Colombia , El Salvador and Mexico , are supporting the case .\nSentence 2: Australia , Chile , Colombia , El Salvador , Honduras , Mexico , New Zealand , Peru and Uruguay will also support the challenge .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is preparing an egg in a skillet\nSentence 2: There is no woman preparing an egg in a skillet", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no woman preparing an egg in a skillet\nSentence 2: A woman is preparing an egg in a skillet", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A small guinea pig is gnawing and eating a piece of carrot on the floor\nSentence 2: There is no small guinea pig gnawing and eating a piece of carrot on the floor", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no small guinea pig gnawing and eating a piece of carrot on the floor\nSentence 2: A small guinea pig is gnawing and eating a piece of carrot on the floor", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 频频什么意思\nSentence 2: 频频是什么意思？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 频频是什么意思？\nSentence 2: 频频什么意思", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is not eating a banana by a tree\nSentence 2: The man is eating a banana by a tree", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is eating a banana by a tree\nSentence 2: The man is not eating a banana by a tree", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你爸爸做什么的\nSentence 2: 打我爸爸", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打我爸爸\nSentence 2: 你爸爸做什么的", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of people in a room, most with computers, some raising their hands.\nSentence 2: Some people have questions.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some people have questions.\nSentence 2: A group of people in a room, most with computers, some raising their hands.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An eggplant is being sliced by a woman\nSentence 2: A woman is cutting a vegetable", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is cutting a vegetable\nSentence 2: An eggplant is being sliced by a woman", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Mrs. Clinton said she was incredulous that he would endanger their marriage and family .\nSentence 2: She hadn 't believed he would jeopardize their marriage and family .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: She hadn 't believed he would jeopardize their marriage and family .\nSentence 2: Mrs. Clinton said she was incredulous that he would endanger their marriage and family .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is stirring vegetables in a pot\nSentence 2: A person is stirring vegetables in a bowl", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is stirring vegetables in a bowl\nSentence 2: A person is stirring vegetables in a pot", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打开同程旅行我要订下周一西安的酒店\nSentence 2: 打开黄瓜视频", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打开黄瓜视频\nSentence 2: 打开同程旅行我要订下周一西安的酒店", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Japan and Russia reportedly have agreed to study the possibility of constructing an oil pipeline connecting Angarsk in eastern Siberia to Russia's Pacific Coast port of Nakhodka.\nSentence 2: An oil pipeline in Siberia was ruptured.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An oil pipeline in Siberia was ruptured.\nSentence 2: Japan and Russia reportedly have agreed to study the possibility of constructing an oil pipeline connecting Angarsk in eastern Siberia to Russia's Pacific Coast port of Nakhodka.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: “粉碎者”于 1987 年 6 月 13 日在日本发行，并由东宝株式会社出版。\nSentence 2: “粉碎者”于 1987 年 6 月 13 日在日本发布，由东宝发行。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: “粉碎者”于 1987 年 6 月 13 日在日本发布，由东宝发行。\nSentence 2: “粉碎者”于 1987 年 6 月 13 日在日本发行，并由东宝株式会社出版。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: At the recent 13th Annual Medicine Meets Virtual Reality Conference in Long Beach, Calif., researchers discussed the development of new technologies using virtual reality to treat soldiers returning from Iraq with posttraumatic stress disorder and to provide those being deployed with stress inoculation training.\nSentence 2: Soldiers are trained using virtual reality.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Soldiers are trained using virtual reality.\nSentence 2: At the recent 13th Annual Medicine Meets Virtual Reality Conference in Long Beach, Calif., researchers discussed the development of new technologies using virtual reality to treat soldiers returning from Iraq with posttraumatic stress disorder and to provide those being deployed with stress inoculation training.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 求大神解答啊\nSentence 2: 求大神解答么么嗒", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 求大神解答么么嗒\nSentence 2: 求大神解答啊", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这张照片是哪个动漫？\nSentence 2: 这张照片是哪个动漫的", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这张照片是哪个动漫的\nSentence 2: 这张照片是哪个动漫？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎样在微信上卖东西？\nSentence 2: 怎么在微信上卖东西", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么在微信上卖东西\nSentence 2: 怎样在微信上卖东西？", "response": "4", "predicted": 4, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is erasing the other man 's work from the board\nSentence 2: The man is erasing the other man 's work from the panel", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is erasing the other man 's work from the panel\nSentence 2: The man is erasing the other man 's work from the board", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 1990年至1992年，Walt Harris在White的Division I-A（现为Division I FBS）Pacific担任相同职位。\nSentence 2: 从1990年到1992年，沃尔特哈里斯在怀特的第一分部（现为第一分部FBS）太平洋分别担任过同样的职务。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 从1990年到1992年，沃尔特哈里斯在怀特的第一分部（现为第一分部FBS）太平洋分别担任过同样的职务。\nSentence 2: 1990年至1992年，Walt Harris在White的Division I-A（现为Division I FBS）Pacific担任相同职位。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天气真好啊，阳光明媚\nSentence 2: 天气真好啊，阳关明媚", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 天气真好啊，阳关明媚\nSentence 2: 天气真好啊，阳光明媚", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Villagers pulling in the fish boat.\nSentence 2: There are people pulling a boat.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are people pulling a boat.\nSentence 2: Villagers pulling in the fish boat.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people, one wearing an orange jersey and a white belt, and another wearing a blue jersey and a red belt are playing flag football.\nSentence 2: Two people on oposite teams play flag football.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people on oposite teams play flag football.\nSentence 2: Two people, one wearing an orange jersey and a white belt, and another wearing a blue jersey and a red belt are playing flag football.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 船在波涛汹涌的水中航行。\nSentence 2: 一个人在波涛汹涌的海面上划皮艇。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个人在波涛汹涌的海面上划皮艇。\nSentence 2: 船在波涛汹涌的水中航行。", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman in a striped shirt holds an infant.\nSentence 2: A woman is dressed.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is dressed.\nSentence 2: A woman in a striped shirt holds an infant.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个日历还有一只蜜蜂是什么成语\nSentence 2: 日历表2015日历一只蜜蜂是什么成语", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 日历表2015日历一只蜜蜂是什么成语\nSentence 2: 一个日历还有一只蜜蜂是什么成语", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOz\nSentence 2: oppogztr", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppogztr\nSentence 2: OPPOz", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Since December 10th, six people have lost their lives in avalanches.\nSentence 2: Humans died in an avalanche.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Humans died in an avalanche.\nSentence 2: Since December 10th, six people have lost their lives in avalanches.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在哪边呢\nSentence 2: 我爸那里", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我爸那里\nSentence 2: 在哪边呢", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy is skillfully playing a piano\nSentence 2: A boy is playing a piano", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A boy is playing a piano\nSentence 2: A boy is skillfully playing a piano", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor7s\nSentence 2: oppor9s", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppor9s\nSentence 2: oppor7s", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 把QQ隐藏起来\nSentence 2: 重新登录QQ", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 重新登录QQ\nSentence 2: 把QQ隐藏起来", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今年的冬天特别冷，尤其是北方地区。\nSentence 2: 今年的冬天特别冷，尤其是北方地去。", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今年的冬天特别冷，尤其是北方地去。\nSentence 2: 今年的冬天特别冷，尤其是北方地区。", "response": "3", "predicted": 3, "true_label": -1}

eval_judge/gpu_results/results_gpu_6.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two dogs run together near the leaves.\nSentence 2: Two dogs are running.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two dogs are running.\nSentence 2: Two dogs run together near the leaves.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Investigators uncovered a 4-inch bone fragment from beneath the concrete slab Thursday , but it turned out to be an animal bone , authorities said .\nSentence 2: Investigators uncovered a 4-inch bone fragment Thursday night , but authorities said it was from an animal .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Investigators uncovered a 4-inch bone fragment Thursday night , but authorities said it was from an animal .\nSentence 2: Investigators uncovered a 4-inch bone fragment from beneath the concrete slab Thursday , but it turned out to be an animal bone , authorities said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The ROE printed here were issued by General Jean Cot, then commander of U.N. forces, and were intended to establish the conditions under which the forces could use their weapons as they carry out the U.N. peacekeeping mission in Bosnia.\nSentence 2: U.N. peacekeeping forces withdrew from Bosnia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: U.N. peacekeeping forces withdrew from Bosnia.\nSentence 2: The ROE printed here were issued by General Jean Cot, then commander of U.N. forces, and were intended to establish the conditions under which the forces could use their weapons as they carry out the U.N. peacekeeping mission in Bosnia.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no person pouring olive oil into a pot on the stove\nSentence 2: A person is pouring olive oil into a pot on the stove", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is pouring olive oil into a pot on the stove\nSentence 2: There is no person pouring olive oil into a pot on the stove", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 求解这是什么动漫！\nSentence 2: 这是什么动漫，求解", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这是什么动漫，求解\nSentence 2: 求解这是什么动漫！", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Microsoft is, undeniably, a company with monopoly power.\nSentence 2: Microsoft holds a monopoly power.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Microsoft holds a monopoly power.\nSentence 2: Microsoft is, undeniably, a company with monopoly power.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 想说什么就说什么\nSentence 2: 说什么好呢", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 说什么好呢\nSentence 2: 想说什么就说什么", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Blond woman overlooking Seattle Space Needle scene.\nSentence 2: Someone taking in the space needle view.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone taking in the space needle view.\nSentence 2: Blond woman overlooking Seattle Space Needle scene.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A group of children divided into a maroon team and a blue team playing a soccer game.\nSentence 2: Some children are playing soccer by a school.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some children are playing soccer by a school.\nSentence 2: A group of children divided into a maroon team and a blue team playing a soccer game.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: hdr\nSentence 2: hdjdjdrhr-duiJ", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: hdjdjdrhr-duiJ\nSentence 2: hdr", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: It indicates , Robert said , “ that terrorists really don ’ t care who they attack .\nSentence 2: \" It also indicates the terrorists really don 't care who they attack . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" It also indicates the terrorists really don 't care who they attack . \"\nSentence 2: It indicates , Robert said , “ that terrorists really don ’ t care who they attack .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国有几个皇帝\nSentence 2: 抗美援朝的纪念日是多少", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 抗美援朝的纪念日是多少\nSentence 2: 中国有几个皇帝", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天去逛街买了一条裙子\nSentence 2: 我昨天去逛街买了一条群子", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我昨天去逛街买了一条群子\nSentence 2: 我昨天去逛街买了一条裙子", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 是的，可以帮您提交受理单，找回支付密码和清空安全保护问题，可以吗\nSentence 2: 这边帮您操作找回登录密码、找回正支付密码、清空安保问题的，您看可以吗？", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这边帮您操作找回登录密码、找回正支付密码、清空安保问题的，您看可以吗？\nSentence 2: 是的，可以帮您提交受理单，找回支付密码和清空安全保护问题，可以吗", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 120秒等于多少分\nSentence 2: 1公里等于多少米", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 1公里等于多少米\nSentence 2: 120秒等于多少分", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A blond little boy in an orange sweatshirt with red sleeves is using scissors to cut something.\nSentence 2: A little male has clothes on with a pair of scissors in his hands.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A little male has clothes on with a pair of scissors in his hands.\nSentence 2: A blond little boy in an orange sweatshirt with red sleeves is using scissors to cut something.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whatcan\nSentence 2: Whatdoes", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whatdoes\nSentence 2: Whatcan", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天去公园散步了，风景真美丽\nSentence 2: 我今天天气散步了，风景真美丽", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气散步了，风景真美丽\nSentence 2: 我今天去公园散步了，风景真美丽", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A worker peers out from atop a building under construction.\nSentence 2: A person is atop of a building.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is atop of a building.\nSentence 2: A worker peers out from atop a building under construction.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and woman are talking\nSentence 2: A man and a woman are silent", "response": "3", "predicted": 3, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are silent\nSentence 2: A man and woman are talking", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 茶壶里煮饺子下一句\nSentence 2: 茶壶里煮饺子的下一句", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 茶壶里煮饺子的下一句\nSentence 2: 茶壶里煮饺子下一句", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 开一下导航\nSentence 2: 开一下电筒", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 开一下电筒\nSentence 2: 开一下导航", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Hampton Township is a few miles northeast of Bay City , about 100 miles away .\nSentence 2: Hampton Township is located a few kilometers northeast of Bay City , near Michigan 's Thumb .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Hampton Township is located a few kilometers northeast of Bay City , near Michigan 's Thumb .\nSentence 2: Hampton Township is a few miles northeast of Bay City , about 100 miles away .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 周三需要了解的10件事\nSentence 2: 周四需要了解的10件事", "response": "1", "predicted": 1, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 周四需要了解的10件事\nSentence 2: 周三需要了解的10件事", "response": "1", "predicted": 1, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 艾肯的研究发表在9月9日的“纽约时报”上。24期“美国医学会杂志”。\nSentence 2: 这一发现发表在周三的“美国医学会杂志”(Journal Of The American Medical Association)上。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这一发现发表在周三的“美国医学会杂志”(Journal Of The American Medical Association)上。\nSentence 2: 艾肯的研究发表在9月9日的“纽约时报”上。24期“美国医学会杂志”。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该物种分布在墨西哥湾和加勒比海;在巴西的大西洋。\nSentence 2: 该物种分布在墨西哥湾和加勒比地区，沿大西洋的巴西。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该物种分布在墨西哥湾和加勒比地区，沿大西洋的巴西。\nSentence 2: 该物种分���在墨西哥湾和加勒比海;在巴西的大西洋。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The earthquake was felt throughout northern Japan, including Hokkaido's largest city, Sapporo, and Kushiro, on the island's eastern shore, but no reports of serious damage have been received.\nSentence 2: An earthquake occurred in Hokkaido.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An earthquake occurred in Hokkaido.\nSentence 2: The earthquake was felt throughout northern Japan, including Hokkaido's largest city, Sapporo, and Kushiro, on the island's eastern shore, but no reports of serious damage have been received.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The dog is playing with a plastic container\nSentence 2: The puppy is not playing with a plastic container", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The puppy is not playing with a plastic container\nSentence 2: The dog is playing with a plastic container", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 三个臭皮匠顶个诸葛亮的意思是什么\nSentence 2: 有关天气的谚语", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有关天气的谚语\nSentence 2: 三个臭皮匠顶个诸葛亮的意思是什么", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Closed sessions are routinely held at the United Nations tribunal that deals with Balkan war crimes , but usually to protect witnesses 's safety .\nSentence 2: Closed sessions are routinely held at the U.N. tribunal that deals with Balkan war crimes , but they are usually closed to protect witnesses who fear for their safety .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Closed sessions are routinely held at the U.N. tribunal that deals with Balkan war crimes , but they are usually closed to protect witnesses who fear for their safety .\nSentence 2: Closed sessions are routinely held at the United Nations tribunal that deals with Balkan war crimes , but usually to protect witnesses 's safety .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 西游记里的人物\nSentence 2: 《西游记》里的人物", "response": "《西游记》里的人物", "predicted": -100, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 《西游记》里的人物\nSentence 2: 西游记里的人物", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Axocomanitla是墨西哥特拉斯卡拉东南部的一个自治市。\nSentence 2: San Lorenzo Axocomanitla是墨西哥东南部特拉斯卡拉的一个自治市。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: San Lorenzo Axocomanitla是墨西哥东南部特拉斯卡拉的一个自治市。\nSentence 2: Axocomanitla是墨西哥特拉斯卡拉东南部的一个自治市。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 1992年12月4日 - 剑桥联合训练师Ian Atkins被任命为伯明翰市教练。\nSentence 2: 1992年12月4日 - 剑桥联队主教练伊恩·阿特金斯被任命为伯明翰市的经理。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 1992年12月4日 - 剑桥联队主教练伊恩·阿特金斯被任命为伯明翰市的经理。\nSentence 2: 1992年12月4日 - 剑桥联合训练师Ian Atkins被任命为伯明翰市教练。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这么看来，白金汉就是拉罗谢尔人的最后一线希望，就是他们的救世主了。\nSentence 2: 白金汉是他们的救世主。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 白金汉是他们的救世主。\nSentence 2: 这么看来，白金汉就是拉罗谢尔人的最后一线希望，就是他们的救世主了。", "response": "4", "predicted": 4, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Chichigalpa的土着人是托尔特克人，Niquiranos和Chorotegas。\nSentence 2: 源自土生土长的Chichigalpa的Toltec人是Niquiranos和Chorotegas。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 源自土生土长的Chichigalpa的Toltec人是Niquiranos和Chorotegas。\nSentence 2: Chichigalpa的土着人是托尔特克人，Niquiranos和Chorotegas。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Weintraub发现了彼得，保罗和玛丽，Randy Newman（他因淫秽而被捕），Lenny Bruce和The Isley Brothers等行为。\nSentence 2: 葡萄发现了彼得，保罗和玛丽，兰迪纽曼（他因淫秽而被捕），莱尼布鲁斯和伊斯利兄弟等行为。", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 葡萄发现了彼得，保罗和玛丽，兰迪纽曼（他因淫秽而被捕），莱尼布鲁斯和伊斯利兄弟等行为。\nSentence 2: Weintraub发现了彼得，保罗和玛丽，Randy Newman（他因淫秽而被捕），Lenny Bruce和The Isley Brothers等行为。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child is walking down the street and a jeep is pulling up\nSentence 2: A child is walking down the street and a jeep is stopping", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child is walking down the street and a jeep is stopping\nSentence 2: A child is walking down the street and a jeep is pulling up", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么本机暂时不支持镜���投屏\nSentence 2: 手机不支持镜像投屏怎么办", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 手机不支持镜像投屏怎么办\nSentence 2: 为什么本机暂时不支持镜像投屏", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 数百名荣誉学生在俄亥俄枪击案中丧生\nSentence 2: 俄亥俄州枪击案受害者的葬礼", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 俄亥俄州枪击案受害者的葬礼\nSentence 2: 数百名荣誉学生在俄亥俄枪击案中丧生", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 迪达勒斯先生从上衣下摆底下掏出烟草袋和烟斗。\nSentence 2: 代达勒斯先生从上衣口袋里掏出了烟袋、烟斗。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 代达勒斯先生从上衣口袋里掏出了烟袋、烟斗。\nSentence 2: 迪达勒斯先生从上衣下摆底下掏出烟草袋和烟斗。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 区内阁由Anne Marit Mevassvik领导，有四名成员，来自保守党，工党和基督教民主党。\nSentence 2: 县内阁由Anne Marit Mevassvik领导，有四名成员，分别来自工党，保守党和基督教民主党。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 县内阁由Anne Marit Mevassvik领导，有四名成员，分别来自工党，保守党和基督教民主党。\nSentence 2: 区内阁由Anne Marit Mevassvik领导，有四名成员，来自保守党，工党和基督教民主党。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 刚刚存储个别图片删除恢复\nSentence 2: 恢复刚刚删的所有图片", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 恢复刚刚删的所有图片\nSentence 2: 刚刚存储个别图片删除恢复", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: “杰里米是个好人，”巴伯说，并补充道：“杰里米正过着这位纽约运动员梦寐以求的生活。”\nSentence 2: 他还说沙基“过着纽约运动员梦寐以求的生活。”", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 他还说沙基“过着纽约运动员梦寐以求的生活。”\nSentence 2: “杰里米是个好人，”巴伯说，并补充道：“杰里米正过着这位纽约运动员梦寐以求的生活。”", "response": "3", "predicted": 3, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么夏天下雨打雷，冬天下雪不打雷？\nSentence 2: 为什么夏天下雨打雷，冬天下雪就不打雷��？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么夏天下雨打雷，冬天下雪就不打雷呢？\nSentence 2: 为什么夏天下雨打雷，冬天下雪不打雷？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这个是一个无效的文本\nSentence 2: 今天的天气不错", "response": "0", "predicted": 0, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天的天气不错\nSentence 2: 这个是一个无效的文本", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A golfer is getting ready to putt on the green, with a crowd of people watching in the background.\nSentence 2: A golfer readies to putt the ball.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A golfer readies to putt the ball.\nSentence 2: A golfer is getting ready to putt on the green, with a crowd of people watching in the background.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Buoyed by some of the advice imparted by Nicklaus , Howell shot an 8-under 64 for a one-stroke lead over Kenny Perry .\nSentence 2: Buoyed by advice imparted by Nicklaus , Howell shot an 8-under 64 on Thursday to enter today 's round with a one-stroke lead over Kenny Perry .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Buoyed by advice imparted by Nicklaus , Howell shot an 8-under 64 on Thursday to enter today 's round with a one-stroke lead over Kenny Perry .\nSentence 2: Buoyed by some of the advice imparted by Nicklaus , Howell shot an 8-under 64 for a one-stroke lead over Kenny Perry .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: chinaunicomvol\nSentence 2: mongolia", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: mongolia\nSentence 2: chinaunicomvol", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: RELAX\nSentence 2: RelaxING", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: RelaxING\nSentence 2: RELAX", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OPPOA79t\nSentence 2: oppoa79ota", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: oppoa79ota\nSentence 2: OPPOA79t", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 怎么没有声音了\nSentence 2: 有什么游戏可以玩吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有什么游戏可以玩吗\nSentence 2: 怎么没有声音了", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个女人和一个男人被耳机逗乐了。\nSentence 2: 一男一女运动后休息。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一男一女运动后休息。\nSentence 2: 一个女人和一个男人被耳机逗乐了。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a red shirt and a blue hat is sitting on a sidewalk and touching drinking glasses.\nSentence 2: a guy with a hat sits with a glass", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a guy with a hat sits with a glass\nSentence 2: A man in a red shirt and a blue hat is sitting on a sidewalk and touching drinking glasses.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The incubation period in cattle is four to five years , said Stephen Sundlof of the U.S. Food and Drug Administration .\nSentence 2: The incubation period in cattle is four to five years , said Dr. Stephen Sundlof of the Food and Drug Administration ( news - web sites ) .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The incubation period in cattle is four to five years , said Dr. Stephen Sundlof of the Food and Drug Administration ( news - web sites ) .\nSentence 2: The incubation period in cattle is four to five years , said Stephen Sundlof of the U.S. Food and Drug Administration .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: YesIam\nSentence 2: Yesyesyes", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Yesyesyes\nSentence 2: YesIam", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 袼劳库斯转身对在座的人说:出卖我的就是这个人，他害得我家破人亡基督教徒们全都知道他的不幸遭遇，维尼茨尤斯也听说过他，不过维尼茨尤斯没有想到他就是格劳库斯，因为这位医生刚才给他包扎伤口时，他已经痛得晕过去了，没有听见他的名字。\nSentence 2: 对戈劳库斯的家史，这里的每个人都非常清楚，只是维尼裘斯不曾想到戈劳库斯就是为他包扎伤口的人，因为当时，他疼得昏死过去，并没有听到这个人的名字。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 对戈劳库斯的家史，这里的每个人都非常清楚，只是维尼裘斯不曾想到戈劳库斯就是为他包扎伤口的人，因为当时，他疼得昏死过去，并没有听到这个人的名字。\nSentence 2: 袼劳库斯转身对在座的人说:出卖我的就是这个人��他害得我家破人亡基督教徒们全都知道他的不幸遭遇，维尼茨尤斯也听说过他，不过维尼茨尤斯没有想到他就是格劳库斯，因为这位医生刚才给他包扎伤口时，他已经痛得晕过去了，没有听见他的名字。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 法国的首都在哪里\nSentence 2: 法国的首都是哪里？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 法国的首都是哪里？\nSentence 2: 法国的首都在哪里", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 鳄鱼害怕什么颜色\nSentence 2: 鳄鱼害怕什么颜色？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 鳄鱼害怕什么颜色？\nSentence 2: 鳄鱼害怕什么颜色", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 发短信给老婆\nSentence 2: 请打开短信", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 请打开短信\nSentence 2: 发短信给老婆", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我要听上海话\nSentence 2: 打开放烟花的声音", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 打开放烟花的声音\nSentence 2: 我要听上海话", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 突触簇形成是指将树突棘添加到通过先前学习添加了其他刺的新区域。\nSentence 2: 突触聚类是指将树突棘添加到通过先前学习添加了其他刺的新区域。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 突触聚类是指将树突棘添加到通过先前学习添加了其他刺的新区域。\nSentence 2: 突触簇形成是指将树突棘添加到通过先前学习添加了其他刺的新区域。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The soccer team, clad in blue for the match, began to counter down the field in front of the defender, clad in red.\nSentence 2: The soccer team in blue plays soccer.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The soccer team in blue plays soccer.\nSentence 2: The soccer team, clad in blue for the match, began to counter down the field in front of the defender, clad in red.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个穿着灰色西装的男人跪在一个穿着白色连衣裙的女人面前。\nSentence 2: 一个穿红衣服的女人和一杯啤酒，后面跟着一个穿白色衣服的男人。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个穿红衣服的女人和一杯啤酒，后面跟着一个穿白色衣服的男人。\nSentence 2: 一个穿着灰色西装的男人跪在一个穿着白色连衣裙的女人面前。", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天我们去动物园玩\nSentence 2: 明天我们去动玩物园", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天我们去动玩物园\nSentence 2: 明天我们去动物园玩", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了午餐\nSentence 2: 我今天吃了误餐", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了误餐\nSentence 2: 我今天吃了午餐", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A clown is singing and people are dancing\nSentence 2: A costumed performer is singing and people are dancing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A costumed performer is singing and people are dancing\nSentence 2: A clown is singing and people are dancing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你知道什么游戏\nSentence 2: 我的游戏呢", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的游戏呢\nSentence 2: 你知道什么游戏", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 与日俱增\nSentence 2: 自作自受的近义词", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 自作自受的近义词\nSentence 2: 与日俱增", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 波多黎各印加省是秘鲁Huánuco地区11个省中最大的省份。\nSentence 2: Puerto Inca省是秘鲁Huánuco地区11个省中最大的省份。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Puerto Inca省是秘鲁Huánuco地区11个省中最大的省份。\nSentence 2: 波多黎各印加省是秘鲁Huánuco地区11个省中最大的省份。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 24的因数有\nSentence 2: 等腰三角形的周长怎么算", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 等腰三角形的周长怎么算\nSentence 2: 24的因数有", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Betty\nSentence 2: PRETTY", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: PRETTY\nSentence 2: Betty", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young woman is shaking hands with a man\nSentence 2: A man and a woman are shaking hands", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a woman are shaking hands\nSentence 2: A young woman is shaking hands with a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is eating some food\nSentence 2: The man is fasting", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is fasting\nSentence 2: A man is eating some food", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A female softball player wearing blue and red crouches in the infield, waiting for the next play.\nSentence 2: a player is waiting", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a player is waiting\nSentence 2: A female softball player wearing blue and red crouches in the infield, waiting for the next play.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The Supreme Court long ago held that students could not be compelled to join in the pledge .\nSentence 2: The U.S. Supreme Court has previously ruled that students are not compelled to say the Pledge of Allegiance .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The U.S. Supreme Court has previously ruled that students are not compelled to say the Pledge of Allegiance .\nSentence 2: The Supreme Court long ago held that students could not be compelled to join in the pledge .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: WHATSYOURFATHER\nSentence 2: Whatsyourhobby", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Whatsyourhobby\nSentence 2: WHATSYOURFATHER", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: sn是什么呀你直接说不就行了\nSentence 2: 要是不说我就认为你是不男不女", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 要是不说我就认为你是不男不女\nSentence 2: sn是什么呀你直接说不就行了", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man in a suit standing at a microphone and singing\nSentence 2: A man in a suit is standing at a microphone and singing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a suit is standing at a microphone and singing\nSentence 2: There is no man in a suit standing at a microphone and singing", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man wearing a short-sleeved blue shirt and carrying a blue backpack while using snow walking sticks treks through the snow with a woman wearing a long-sleeved blue shirt and black pants also using snow walking sticks.\nSentence 2: a guy with a blue top carries a pack and makes his way through snow", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a guy with a blue top carries a pack and makes his way through snow\nSentence 2: A man wearing a short-sleeved blue shirt and carrying a blue backpack while using snow walking sticks treks through the snow with a woman wearing a long-sleeved blue shirt and black pants also using snow walking sticks.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 男人带什么项链好\nSentence 2: 男人带什么玉项链好", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 男人带什么玉项链好\nSentence 2: 男人带什么项链好", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: After Saddam 's regime crumbled in early April , it had to wait for legal hurdles to be crossed before sales could resume .\nSentence 2: After Saddam 's regime crumbled in early April , legal hurdles had to be cleared before sales could resume .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: After Saddam 's regime crumbled in early April , legal hurdles had to be cleared before sales could resume .\nSentence 2: After Saddam 's regime crumbled in early April , it had to wait for legal hurdles to be crossed before sales could resume .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喜欢在下半场跑比第一节快1-2分钟。\nSentence 2: 我肯定会在上半场稍微慢一点。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我肯定会在上半场稍微慢一点。\nSentence 2: 我喜欢在下半场跑比第一节快1-2分钟。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 孟买是印度第二大生活成本最高的城市。\nSentence 2: 印度科技之都班加罗尔是印度生活成本最高的城市：调查", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 印度科技之都班加罗尔是印度生活成本最高的城市：调查\nSentence 2: 孟买是印度第二大生活成本最高的城市。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: CHINESE\nSentence 2: Chinesebook", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Chinesebook\nSentence 2: CHINESE", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你喜欢范丞丞吗\nSentence 2: 王者荣耀有多少个英雄", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 王者荣耀有多少个英雄\nSentence 2: 你喜欢范丞丞吗", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is not kick boxing with a trainer\nSentence 2: The man is kick boxing with a trainer", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is kick boxing with a trainer\nSentence 2: The man is not kick boxing with a trainer", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么东西越热它爬的越高。\nSentence 2: 什么东西越热爬的越高", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么东西越热爬的越高\nSentence 2: 什么东西越热它爬的越高。", "response": "3", "predicted": 3, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你会说什么\nSentence 2: 你会说白话吗", "response": "1", "predicted": 1, "true_label": 1}

eval_judge/gpu_results/results_gpu_7.jsonl

@@ -0,0 +1,175 @@
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你会说白话吗\nSentence 2: 你会说什么", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 埃及逮捕穆斯林兄弟会领导人巴迪\nSentence 2: 埃及准备应对更多的暴力，穆斯林兄弟会称其为“愤怒的游行”。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 埃及准备应对更多的暴力，穆斯林兄弟会称其为“愤怒的游行”。\nSentence 2: 埃及逮捕穆斯林兄弟会领导人巴迪", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: They came despite what BA called a \" difficult quarter \" , which it said included unofficial industrial action at Heathrow .\nSentence 2: BA said the second quarter , which included unofficial industrial action at Heathrow , had been difficult .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: BA said the second quarter , which included unofficial industrial action at Heathrow , had been difficult .\nSentence 2: They came despite what BA called a \" difficult quarter \" , which it said included unofficial industrial action at Heathrow .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A woman is holding her child while a man is reading something from a piece of paper and taking something from a bowl that a younger girl is holding in a park with a small lake in the background.\nSentence 2: There are people outside", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There are people outside\nSentence 2: A woman is holding her child while a man is reading something from a piece of paper and taking something from a bowl that a younger girl is holding in a park with a small lake in the background.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了午饭\nSentence 2: 我今天天气吃了", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天天气吃了\nSentence 2: 我今天吃了午饭", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Nickerson、Duquette、Bruno、Askov、Sandstone 和 Sturgeon Lake 社区都在克里克附近。\nSentence 2: 布鲁诺、度凯特、尼克森、阿斯科夫、桑德斯东和特金莱克社区都离克里克很近。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 布鲁诺、度凯特、尼克森、阿斯科夫、桑德斯东和特金莱克社区都离克里克很近。\nSentence 2: Nickerson、Duquette、Bruno、Askov、Sandstone 和 Sturgeon Lake 社区都在克里克附近。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The Longuet and Leotard Affairs (1996) concern both personal privileges for, and illegal financing of, the Republican Party leaders.\nSentence 2: A French political party was illegally financed.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A French political party was illegally financed.\nSentence 2: The Longuet and Leotard Affairs (1996) concern both personal privileges for, and illegal financing of, the Republican Party leaders.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The rhino is grazing on the grass\nSentence 2: A animal is grazing in a field", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A animal is grazing in a field\nSentence 2: The rhino is grazing on the grass", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么牌子的净水器好\nSentence 2: 什么牌子净水器好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么牌子净水器好\nSentence 2: 什么牌子的净水器好", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 查询别人手机话费余额\nSentence 2: 如何在网上查询别人手机话费余额", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 如何在网上查询别人手机话费余额\nSentence 2: 查询别人手机话费余额", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: pccm 00\nSentence 2: OTC", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OTC\nSentence 2: pccm 00", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有什么好看的恐怖电影吗？\nSentence 2: 有什么好看的恐怖电影吗?", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 有什么好看的恐怖电影吗?\nSentence 2: 有什么好看的恐怖电影吗？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Workers standing on a lift.\nSentence 2: Workers stand on a lift", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Workers stand on a lift\nSentence 2: Workers standing on a lift.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 二十分钟倒计时开始是几时\nSentence 2: 我想要二十分钟提醒我", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我想要二十分钟提醒我\nSentence 2: 二十分钟倒计时开始是几时", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A busy street with numerous people interacting and going about their lives.\nSentence 2: The street is alive with activity.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The street is alive with activity.\nSentence 2: A busy street with numerous people interacting and going about their lives.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man with a tan jacket with a full grocery bag is crossing the street.\nSentence 2: The man just went shopping at the grocery store.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man just went shopping at the grocery store.\nSentence 2: A man with a tan jacket with a full grocery bag is crossing the street.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Scp96\nSentence 2: scp001", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: scp001\nSentence 2: Scp96", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 两个村庄位于波蒂奇（Portage）：南部的杰里市（Jerry City）和西北部的波蒂奇（Portage）镇的一部分。\nSentence 2: 在波蒂奇镇有两个村庄：南部的杰瑞城的一部分和西北部的波蒂奇的一部分。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在波蒂奇镇有两个村庄：南部的杰瑞城的一部分和西北部的波蒂奇的一部分。\nSentence 2: 两个村庄位于波蒂奇（Portage）：南部的杰里市（Jerry City）和西北部的波蒂奇（Portage）镇的一部分。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 然而，格鲁吉亚语在指定数量时使用单数形式，因此在实践中，“tetri”的复数只是“tetri”。\nSentence 2: 然而，当指定数量时，格鲁吉亚语言是单数形式，因此在实践中，“tetri”的复数仅使用“tetri”。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 然而，当指定数量时，格鲁吉亚语言是单数形式，因此在实践中，“tetri”的复数仅使用“tetri”。\nSentence 2: 然而，格鲁吉亚语在指定数量时使用单数形式，因此在实践中，“tetri”的复数只是“tetri”。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 开船的时候，头里就觉得极不舒服。\nSentence 2: 开船的时候，他就感到头疼��", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 开船的时候，他就感到头疼。\nSentence 2: 开船的时候，头里就觉得极不舒服。", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么星座的女生最漂亮\nSentence 2: 什么星座的女生最漂亮？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 什么星座的女生最漂亮？\nSentence 2: 什么星座的女生最漂亮", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我喝酸奶你舔酸奶盖盖\nSentence 2: 西红柿炒鸡蛋的做法", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 西红柿炒鸡蛋的做法\nSentence 2: 我喝酸奶你舔酸奶盖盖", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么托马斯的书被排除在正典之外？\nSentence 2: 为什么会有更多的耶稣话语被排除在正典之外呢？", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么会有更多的耶稣话语被排除在正典之外呢？\nSentence 2: 为什么托马斯的书被排除在正典之外？", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is singing and playing a guitar\nSentence 2: A man is playing a guitar", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is playing a guitar\nSentence 2: A man is singing and playing a guitar", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个人正在把胡萝卜切成碎片。\nSentence 2: 一个人正在把洋葱切成小块。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 一个人正在把洋葱切成小块。\nSentence 2: 一个人正在把胡萝卜切成碎片。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: To protect Tibet's ecological environment, the government exercises a \"felling by quota\" policy, and strictly controls the scale of tree-felling in forests.\nSentence 2: The government took measures to protect the environment.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The government took measures to protect the environment.\nSentence 2: To protect Tibet's ecological environment, the government exercises a \"felling by quota\" policy, and strictly controls the scale of tree-felling in forests.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A young boy in red leaping into sand at a playground.\nSentence 2: A child is playing in the sand.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A child is playing in the sand.\nSentence 2: A young boy in red leaping into sand at a playground.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A dog is eating a doll\nSentence 2: A dog is biting a doll", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A dog is biting a doll\nSentence 2: A dog is eating a doll", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我做梦梦里有人在打麻将\nSentence 2: 王者荣耀里有哪些歌", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 王者荣耀里有哪些歌\nSentence 2: 我做梦梦里有人在打麻将", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An 11-year-old Sanford boy was bitten on the foot by a shark.\nSentence 2: A shark attacked a human being.", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A shark attacked a human being.\nSentence 2: An 11-year-old Sanford boy was bitten on the foot by a shark.", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A mother and daughter walk along the side of a bridge.\nSentence 2: A mother and daughter are walking.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A mother and daughter are walking.\nSentence 2: A mother and daughter walk along the side of a bridge.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is showing off a small monkey\nSentence 2: A man is holding an animal , which is small , in one hand", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is holding an animal , which is small , in one hand\nSentence 2: A man is showing off a small monkey", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 根据美国人口普查局的统计，该县的总面积为土地和（0.2％）水。\nSentence 2: 根据美国人口普查局的统计，该县的总面积为陆地，（0.2％）为水。", "response": "5", "predicted": 5, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 根据美国人口普查局的统计，该���的总面积为陆地，（0.2％）为水。\nSentence 2: 根据美国人口普查局的统计，该县的总面积为土地和（0.2％）水。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Exports of cars, which used to be the main product, had by then decreased to 250 million euro.\nSentence 2: Cars exported by Japan decreased.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cars exported by Japan decreased.\nSentence 2: Exports of cars, which used to be the main product, had by then decreased to 250 million euro.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么手机没有声音了\nSentence 2: 为什么打游戏没声音玩别的应用却有声音", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 为什么打游戏没声音玩别的应用却有声音\nSentence 2: 为什么手机没有声音了", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A blond woman in a black shirt is standing behind a counter.\nSentence 2: The woman is standing.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The woman is standing.\nSentence 2: A blond woman in a black shirt is standing behind a counter.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is praying fervently\nSentence 2: A man is praying", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is praying\nSentence 2: A man is praying fervently", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Magner , who is 54 and known as Marge , has been the consumer group 's chief operating officer since April 2002 , and sits on Citigroup 's management committee .\nSentence 2: She has been the consumer unit 's chief operating officer since April 2002 , and sits on Citigroup 's management committee .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: She has been the consumer unit 's chief operating officer since April 2002 , and sits on Citigroup 's management committee .\nSentence 2: Magner , who is 54 and known as Marge , has been the consumer group 's chief operating officer since April 2002 , and sits on Citigroup 's management committee .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man playing a bamboo flute\nSentence 2: A man is playing a bamboo flute", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is playing a bamboo flute\nSentence 2: There is no man playing a bamboo flute", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 印度尼西亚岛外发生6.3级地震\nSentence 2: 泰国发生6级地震", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 泰国发生6级地震\nSentence 2: 印度尼西亚岛外发生6.3级地震", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is painting a picture outside behind a crowd.\nSentence 2: A painter is creating a picture.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A painter is creating a picture.\nSentence 2: A man is painting a picture outside behind a crowd.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is sitting in front of a graffiti covered wall.\nSentence 2: There's a place to sit near a wall", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There's a place to sit near a wall\nSentence 2: A person is sitting in front of a graffiti covered wall.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The military officer is barking at the recruits\nSentence 2: An officer is talking to the soldiers", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An officer is talking to the soldiers\nSentence 2: The military officer is barking at the recruits", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我要听娱乐新闻\nSentence 2: 新闻早播报", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 新闻早播报\nSentence 2: 我要听娱乐新闻", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在微信中充三十元话费\nSentence 2: 微信充话费也很方便", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 微信充话费也很方便\nSentence 2: 在微信中充三十元话费", "response": "-1", "predicted": -1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该党开始是一场抵抗运动，为东帝汶在1974年至1998年之间的独立而战，首先是葡萄牙，然后是印度尼西亚。\nSentence 2: 该党始于1974年至1998年期间首次从印度尼西亚和葡萄牙争取东帝汶独立的抵抗运动。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 该党始于1974年至1998年期间首次从印度尼西亚和葡萄牙争取东帝汶独立的抵抗运动���\nSentence 2: 该党开始是一场抵抗运动，为东帝汶在1974年至1998年之间的独立而战，首先是葡萄牙，然后是印度尼西亚。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 如果一项声明不能证明，则可能或将会出现使系统无效的不一致或自相矛盾之处。\nSentence 2: 在这一点上的到来是否应该成为对任何系统进行检查的重点？", "response": "0", "predicted": 0, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 在这一点上的到来是否应该成为对任何系统进行检查的重点？\nSentence 2: 如果一项声明不能证明，则可能或将会出现使系统无效的不一致或自相矛盾之处。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气不错吗吗吗\nSentence 2: 今天天气好好好好", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气好好好好\nSentence 2: 今天天气不错吗吗吗", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 空间访问量怎么刷啊？\nSentence 2: 空间访问量怎么刷啊。要速度快的！", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 空间访问量怎么刷啊。要速度快的！\nSentence 2: 空间访问量怎么刷啊？", "response": "5", "predicted": 5, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 定个十点半的闹铃\nSentence 2: 设个五点钟的闹钟", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 设个五点钟的闹钟\nSentence 2: 定个十点半的闹铃", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: If destruction of the rainforest continues, global warming will continue to take place.\nSentence 2: Destruction of the rainforest affects the world's weather.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Destruction of the rainforest affects the world's weather.\nSentence 2: If destruction of the rainforest continues, global warming will continue to take place.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 吃巧克力吗\nSentence 2: 孕妇可以吃巧克力吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 孕妇可以吃巧克力吗\nSentence 2: 吃巧克力吗", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Aiken 's study appears in the Sept . 24 issue of the Journal of the American Medical Association .\nSentence 2: The findings appear in Wednesday 's Journal of the American Medical Association .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The findings appear in Wednesday 's Journal of the American Medical Association .\nSentence 2: Aiken 's study appears in the Sept . 24 issue of the Journal of the American Medical Association .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 七夕节怎么过！哎\nSentence 2: 七夕节怎么过才好", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 七夕节怎么过才好\nSentence 2: 七夕节怎么过！哎", "response": "1", "predicted": 1, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OKOKOK\nSentence 2: OKTHANKYOU", "response": "4", "predicted": 4, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: OKTHANKYOU\nSentence 2: OKOKOK", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man in a plaid shirt looking through a telescope lens.\nSentence 2: a man is looking through a telescope", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: a man is looking through a telescope\nSentence 2: A man in a plaid shirt looking through a telescope lens.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我说让你滚听不懂人话吗\nSentence 2: 我的手机，说话听不见", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我的手机，说话听不见\nSentence 2: 我说让你滚听不懂人话吗", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people talking on a dock.\nSentence 2: people outside", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: people outside\nSentence 2: Two people talking on a dock.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 她叔不会讨厌抽烟吧？\nSentence 2: 你不会讨厌抽烟吧？", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你不会讨厌抽烟吧？\nSentence 2: 她叔不会讨厌抽烟吧？", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two ladies hiding behind their cotton candy, and smiling on the sidewalk.\nSentence 2: Two friends holding cotton candy smile at each other.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two friends holding cotton candy smile at each other.\nSentence 2: Two ladies hiding behind their cotton candy, and smiling on the sidewalk.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An medical worker examines a young girl.\nSentence 2: A health worker is evaluating a girl on her sickness.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A health worker is evaluating a girl on her sickness.\nSentence 2: An medical worker examines a young girl.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我想去公园散步\nSentence 2: 今天天气真好，我想去公园散b散步", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 今天天气真好，我想去公园散b散步\nSentence 2: 今天天气真好，我想去公园散步", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three men are carrying a red bag into a boat with another person and boat in the background.\nSentence 2: Some people put something in a boat in a place with more than one boat.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some people put something in a boat in a place with more than one boat.\nSentence 2: Three men are carrying a red bag into a boat with another person and boat in the background.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 中国西南部道路交通事故10人死亡5人受伤\nSentence 2: 华东公路交通事故死亡1人39人受伤", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 华东公路交通事故死亡1人39人受伤\nSentence 2: 中国西南部道路交通事故10人死亡5人受伤", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The number eight rollerskater is taking the lead and pulling away from her black and red dressed opponents.\nSentence 2: A skater ahead of the competition in a race.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A skater ahead of the competition in a race.\nSentence 2: The number eight rollerskater is taking the lead and pulling away from her black and red dressed opponents.", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: His band in the early 1930 's included the pianist Teddy Wilson , the saxophonist Chu Berry , the trombonist J. C. Higginbotham and the drummer Sid Catlett .\nSentence 2: His band in the early 1930s included pianist Teddy Wilson , saxophonist Chu Berry , trombonist J.C. Higginbotham and drummer Sid Catlett .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: His band in the early 1930s included pianist Teddy Wilson , saxophonist Chu Berry , trombonist J.C. Higginbotham and drummer Sid Catlett .\nSentence 2: His band in the early 1930 's included the pianist Teddy Wilson , the saxophonist Chu Berry , the trombonist J. C. Higginbotham and the drummer Sid Catlett .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: An arrest warrant claimed Bryant assaulted the woman June 30 at a hotel .\nSentence 2: According to an arrest warrant , Bryant , 24 , attacked a woman on June 30 .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: According to an arrest warrant , Bryant , 24 , attacked a woman on June 30 .\nSentence 2: An arrest warrant claimed Bryant assaulted the woman June 30 at a hotel .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 科比尼切夫的实验室与美国劳伦斯利弗莫尔国家实验室。\nSentence 2: 科比尼切夫的实验室与其他科学机构合作。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 科比尼切夫的实验室与其他科学机构合作。\nSentence 2: 科比尼切夫的实验室与美国劳伦斯利弗莫尔国家实验室。", "response": "2", "predicted": 2, "true_label": 2}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is eating a pizza\nSentence 2: Some food is being eaten by a man", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Some food is being eaten by a man\nSentence 2: A man is eating a pizza", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 此外，西班牙与葡萄牙法院进行谈判，并同意于1668年2月13日实现里斯本和平。\nSentence 2: 此外，西班牙同意与葡萄牙法院进行谈判，并于1668年2月13日进入里斯本和平。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 此外，西班牙同意与葡萄牙法院进行谈判，并于1668年2月13日进入里斯本和平。\nSentence 2: 此外，西班牙与葡萄牙法院进行谈判，并同意于1668年2月13日实现里斯本和平。", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is cutting pieces of butter into a mixing bowl\nSentence 2: Pieces of butter are being added to the stand mixer", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Pieces of butter are being added to the stand mixer\nSentence 2: A man is cutting pieces of butter into a mixing bowl", "response": "3", "predicted": 3, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: RealMe手机如何返修\nSentence 2: RealMe手机维修一般要多久", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: RealMe手机维修一般要多久\nSentence 2: RealMe手机如何返修", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is dancing\nSentence 2: A person is dancing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A person is dancing\nSentence 2: A man is dancing", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man and a child are purchasing ice cream from a man with a hand pushed ice cream cart.\nSentence 2: People are buying food from a street vendor.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: People are buying food from a street vendor.\nSentence 2: A man and a child are purchasing ice cream from a man with a hand pushed ice cream cart.", "response": "3", "predicted": 3, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Someone is drilling a hole in a strip of wood with a power drill\nSentence 2: A man is drilling a hole in a piece of wood", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A man is drilling a hole in a piece of wood\nSentence 2: Someone is drilling a hole in a strip of wood with a power drill", "response": "4", "predicted": 4, "true_label": 4}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three workers, with life vests and hard hats, on a boat hanging from cables over water.\nSentence 2: Three people are on a boat.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Three people are on a boat.\nSentence 2: Three workers, with life vests and hard hats, on a boat hanging from cables over water.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: \" I felt that if I disagreed with Rosie too much I would lose my job , \" she said .\nSentence 2: Cavender did say : \" I felt that if I disagreed with Rosie too much I would lose my job . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cavender did say : \" I felt that if I disagreed with Rosie too much I would lose my job . \"\nSentence 2: \" I felt that if I disagreed with Rosie too much I would lose my job , \" she said .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了午饭\nSentence 2: 我今天吃了午fan", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 我今天吃了午fan\nSentence 2: 我今天吃了午饭", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Slightly more than half of the profit shortfall results from a sales slump , with weakness spread across the company 's various geographic and end markets .\nSentence 2: Slightly more than half of the earnings miss was due to a sales slump , with weakness was spread across the company 's various geographic and end markets .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Slightly more than half of the earnings miss was due to a sales slump , with weakness was spread across the company 's various geographic and end markets .\nSentence 2: Slightly more than half of the profit shortfall results from a sales slump , with weakness spread across the company 's various geographic and end markets .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Other features include FileVault , which secures the contents of a home directory with 128-bit AES encryption on the fly .\nSentence 2: A new feature dubbed FileVault , also new in Panther , secures the contents of a user 's home directory with 128-bit AES encryption .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A new feature dubbed FileVault , also new in Panther , secures the contents of a user 's home directory with 128-bit AES encryption .\nSentence 2: Other features include FileVault , which secures the contents of a home directory with 128-bit AES encryption on the fly .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: According to the \"Current Production Statistics Survey\" of METI, the index of industrial production of Japan in 2002 decreased 1.4% from the previous year to 91.9.\nSentence 2: Cars exported by Japan decreased.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Cars exported by Japan decreased.\nSentence 2: According to the \"Current Production Statistics Survey\" of METI, the index of industrial production of Japan in 2002 decreased 1.4% from the previous year to 91.9.", "response": "-1", "predicted": -1, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 谢霆锋和张柏芝复婚吗？\nSentence 2: 谢霆锋和张柏芝复婚了吗", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 谢霆锋和张柏芝复婚了吗\nSentence 2: 谢霆锋和张柏芝复婚吗？", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你能不能给我一百万\nSentence 2: 你能给我发个网站吗", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 你能给我发个网站吗\nSentence 2: 你能不能给我一百万", "response": "1", "predicted": 1, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The American decision provoked an angry reaction from the European Commission , which described the move as \" legally unwarranted , economically unfounded and politically unhelpful \" .\nSentence 2: The European Commission , the EU 's powerful executive body , described the move as \" legally unwarranted , economically unfounded and politically unhelpful . \"", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The European Commission , the EU 's powerful executive body , described the move as \" legally unwarranted , economically unfounded and politically unhelpful . \"\nSentence 2: The American decision provoked an angry reaction from the European Commission , which described the move as \" legally unwarranted , economically unfounded and politically unhelpful \" .", "response": "5", "predicted": 5, "true_label": 5}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: There is no man singing and playing the guitar\nSentence 2: The man is singing and playing the guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The man is singing and playing the guitar\nSentence 2: There is no man singing and playing the guitar", "response": "0", "predicted": 0, "true_label": 0}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 明天后天下不下雨\nSentence 2: 这一周的天气", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: 这一周的天气\nSentence 2: 明天后天下不下雨", "response": "1", "predicted": 1, "true_label": 1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Each year the federal government purchases a large volume of food and commodities for its domestic, international, and military programs.\nSentence 2: A government purchases food.", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: A government purchases food.\nSentence 2: Each year the federal government purchases a large volume of food and commodities for its domestic, international, and military programs.", "response": "3", "predicted": 3, "true_label": -1}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: Two people crossing by each other while kite surfing.\nSentence 2: The people are outdoors.", "response": "2", "predicted": 2, "true_label": 3}
+{"prompt": "Please rate the semantic similarity of the following two sentences on a scale from -1 to 5, where 1 means completely different and 5 means completely equivalent in meaning. Only output the number.\nSentence 1: The people are outdoors.\nSentence 2: Two people crossing by each other while kite surfing.", "response": "2", "predicted": 2, "true_label": 3}

eval_judge/metrics.csv

@@ -0,0 +1,8 @@
+Metric,Value
+Accuracy,0.8043
+Precision_Weighted,0.8112
+Recall_Weighted,0.8043
+F1_Weighted,0.8033
+Precision_Macro,0.7098
+Recall_Macro,0.7038
+F1_Macro,0.7029

eval_judge/metrics.json

@@ -0,0 +1,74 @@
+{
+    "accuracy": 0.8042857142857143,
+    "precision_weighted": 0.8112312499728647,
+    "recall_weighted": 0.8042857142857143,
+    "f1_score_weighted": 0.8032805197335097,
+    "precision_macro": 0.7098273437262567,
+    "recall_macro": 0.7037500000000001,
+    "f1_score_macro": 0.702870454766821,
+    "confusion_matrix": [
+        [
+            186,
+            1,
+            1,
+            0,
+            9,
+            0,
+            3
+        ],
+        [
+            0,
+            163,
+            35,
+            0,
+            2,
+            0,
+            0
+        ],
+        [
+            1,
+            0,
+            188,
+            2,
+            0,
+            9,
+            0
+        ],
+        [
+            0,
+            9,
+            3,
+            149,
+            39,
+            0,
+            0
+        ],
+        [
+            3,
+            0,
+            39,
+            44,
+            101,
+            7,
+            6
+        ],
+        [
+            0,
+            0,
+            0,
+            0,
+            20,
+            156,
+            24
+        ],
+        [
+            2,
+            0,
+            6,
+            0,
+            2,
+            5,
+            183
+        ]
+    ]
+}

eval_judge/metrics_transposed.csv

@@ -0,0 +1,2 @@
+Metric,Accuracy,Precision_Weighted,Recall_Weighted,F1_Weighted,Precision_Macro,Recall_Macro,F1_Macro
+Value,0.8043,0.8112,0.8043,0.8033,0.7098,0.7038,0.7029

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 35.0,
+    "eval_loss": 2.5913538932800293,
+    "eval_runtime": 0.4619,
+    "eval_samples_per_second": 303.127,
+    "eval_steps_per_second": 2.165
+}

generation_config.json

@@ -0,0 +1,8 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "temperature": 0.8,
+  "top_p": 0.8,
+  "transformers_version": "4.52.1"
+}

model-00001-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db149071b4771bdd8ef09fd2125aaafe7d186e8614c2ba96859a1594f797f740
+size 4993234616

model-00002-of-00002.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8492fa93882aebd87b55a0512cfb3c8025d78ddd1541667e396c86d6573c9be0
+size 1022215072

model.safetensors.index.json

@@ -0,0 +1,370 @@
+{
+  "metadata": {
+    "total_size": 6015407616
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00002-of-00002.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00002-of-00002.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00002.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00002.safetensors",
+    "model.norm.weight": "model-00002-of-00002.safetensors"
+  }
+}

modeling_minicpm.py

@@ -0,0 +1,1453 @@
+# coding=utf-8
+# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
+#
+# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
+# and OPT implementations in this library. It has been modified from its
+# original forms to accommodate minor architectural differences compared
+# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+""" PyTorch MiniCPM model."""
+import math
+import warnings
+from typing import List, Optional, Tuple, Union, Dict
+
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from torch import nn
+from torch.nn import BCEWithLogitsLoss, CrossEntropyLoss, MSELoss
+
+from transformers.activations import ACT2FN
+from transformers.cache_utils import Cache, DynamicCache
+from transformers.modeling_attn_mask_utils import (
+    AttentionMaskConverter,
+    _prepare_4d_attention_mask,
+    _prepare_4d_causal_attention_mask,
+    _prepare_4d_causal_attention_mask_for_sdpa,
+)
+from transformers.modeling_outputs import BaseModelOutputWithPast, CausalLMOutputWithPast, SequenceClassifierOutputWithPast
+from transformers.modeling_utils import PreTrainedModel
+from transformers.pytorch_utils import ALL_LAYERNORM_LAYERS, is_torch_greater_or_equal_than_1_13
+from transformers.utils import (
+    add_start_docstrings,
+    add_start_docstrings_to_model_forward,
+    is_flash_attn_2_available,
+    is_flash_attn_greater_or_equal_2_10,
+    logging,
+    replace_return_docstrings,
+)
+from transformers.utils.import_utils import is_torch_fx_available
+from .configuration_minicpm import MiniCPMConfig
+import re
+
+try:
+    from flash_attn import flash_attn_func, flash_attn_varlen_func
+    from flash_attn.bert_padding import index_first_axis, pad_input, unpad_input  # noqa
+except:
+    pass
+
+
+# This makes `_prepare_4d_causal_attention_mask` a leaf function in the FX graph.
+# It means that the function will not be traced through and simply appear as a node in the graph.
+if is_torch_fx_available():
+    if not is_torch_greater_or_equal_than_1_13:
+        import torch.fx
+
+    _prepare_4d_causal_attention_mask = torch.fx.wrap(_prepare_4d_causal_attention_mask)
+
+
+logger = logging.get_logger(__name__)
+
+_CONFIG_FOR_DOC = "MiniCPMConfig"
+
+
+def _get_unpad_data(attention_mask):
+    seqlens_in_batch = attention_mask.sum(dim=-1, dtype=torch.int32)
+    indices = torch.nonzero(attention_mask.flatten(), as_tuple=False).flatten()
+    max_seqlen_in_batch = seqlens_in_batch.max().item()
+    cu_seqlens = F.pad(torch.cumsum(seqlens_in_batch, dim=0, dtype=torch.torch.int32), (1, 0))
+    return (
+        indices,
+        cu_seqlens,
+        max_seqlen_in_batch,
+    )
+
+
+def _expand_mask(mask: torch.Tensor, dtype: torch.dtype, tgt_len: Optional[int] = None):
+    warnings.warn(
+        "Calling `transformers.models.minicpm.modeling_minicpm._prepare_4d_attention_mask` is deprecated and will be removed in v4.37. Use `transformers.modeling_attn_mask_utils._prepare_4d_attention_mask"
+    )
+    return _prepare_4d_attention_mask(mask=mask, dtype=dtype, tgt_len=tgt_len)
+
+
+def _make_causal_mask(
+    input_ids_shape: torch.Size, dtype: torch.dtype, device: torch.device, past_key_values_length: int = 0
+):
+    warnings.warn(
+        "Calling `transformers.models.minicpm.modeling_minicpm._make_causal_mask` is deprecated and will be removed in v4.37. Use `transformers.models.minicpm.modeling_minicpm.AttentionMaskConverter._make_causal_mask"
+    )
+    return AttentionMaskConverter._make_causal_mask(
+        input_ids_shape=input_ids_shape, dtype=dtype, device=device, past_key_values_length=past_key_values_length
+    )
+
+# @torch.jit.script  # type: ignore
+def rms_layernorm(hidden: torch.Tensor, weight: torch.Tensor, eps: float):
+    old_dtype = hidden.dtype
+    variance = hidden.to(torch.float32).pow(2).mean(dim=-1, keepdim=True)
+    hidden = (hidden * torch.rsqrt(variance + eps)).to(old_dtype)
+    return hidden * weight
+
+
+class MiniCPMRMSNorm(nn.Module):
+    def __init__(self, hidden_size, eps=1e-6):
+        """
+        MiniCPMRMSNorm is equivalent to T5LayerNorm
+        """
+        super().__init__()
+        self.weight = nn.Parameter(torch.ones(hidden_size))
+        self.variance_epsilon = eps
+
+    def forward(self, hidden_states):
+        return rms_layernorm(hidden_states, self.weight, self.variance_epsilon)
+
+
+ALL_LAYERNORM_LAYERS.append(MiniCPMRMSNorm)
+
+
+class MiniCPMRotaryEmbedding(nn.Module):
+    def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None):
+        super().__init__()
+
+        self.dim = dim
+        self.max_position_embeddings = max_position_embeddings
+        self.base = base
+        inv_freq = 1.0 / (self.base ** (torch.arange(0, self.dim, 2).float().to(device) / self.dim))
+        self.register_buffer("inv_freq", inv_freq, persistent=False)
+
+        # Build here to make `torch.jit.trace` work.
+        self._set_cos_sin_cache(
+            # seq_len=max_position_embeddings, device=self.inv_freq.device, dtype=torch.get_default_dtype()
+            seq_len=max_position_embeddings, device=self.inv_freq.device, dtype=torch.float32
+        )
+
+    def _set_cos_sin_cache(self, seq_len, device, dtype):
+        self.max_seq_len_cached = seq_len
+        t = torch.arange(self.max_seq_len_cached, device=device, dtype=self.inv_freq.dtype)
+        freqs = torch.outer(t, self.inv_freq)
+        # Different from paper, but it uses a different permutation in order to obtain the same calculation
+        emb = torch.cat((freqs, freqs), dim=-1)
+
+        self.register_buffer("cos_cached", emb.cos().to(dtype), persistent=False)
+        self.register_buffer("sin_cached", emb.sin().to(dtype), persistent=False)
+
+    def forward(self, x, seq_len=None):
+        # x: [bs, num_attention_heads, seq_len, head_size]
+        if seq_len > self.max_seq_len_cached:
+            self._set_cos_sin_cache(seq_len=seq_len, device=x.device, dtype=x.dtype)
+
+        return (
+            self.cos_cached[:seq_len].to(dtype=x.dtype),
+            self.sin_cached[:seq_len].to(dtype=x.dtype),
+        )
+
+
+class MiniCPMLinearScalingRotaryEmbedding(MiniCPMRotaryEmbedding):
+    """MiniCPMRotaryEmbedding extended with linear scaling. Credits to the Reddit user /u/kaiokendev"""
+
+    def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None, scaling_factor=1.0):
+        self.scaling_factor = scaling_factor
+        super().__init__(dim, max_position_embeddings, base, device)
+
+    def _set_cos_sin_cache(self, seq_len, device, dtype):
+        self.max_seq_len_cached = seq_len
+        t = torch.arange(self.max_seq_len_cached, device=device, dtype=self.inv_freq.dtype)
+        t = t / self.scaling_factor
+
+        freqs = torch.outer(t, self.inv_freq)
+        # Different from paper, but it uses a different permutation in order to obtain the same calculation
+        emb = torch.cat((freqs, freqs), dim=-1)
+        self.register_buffer("cos_cached", emb.cos().to(dtype), persistent=False)
+        self.register_buffer("sin_cached", emb.sin().to(dtype), persistent=False)
+
+
+class MiniCPMDynamicNTKScalingRotaryEmbedding(MiniCPMRotaryEmbedding):
+    """MiniCPMRotaryEmbedding extended with Dynamic NTK scaling. Credits to the Reddit users /u/bloc97 and /u/emozilla"""
+
+    def __init__(self, dim, max_position_embeddings=2048, base=10000, device=None, scaling_factor=1.0):
+        self.scaling_factor = scaling_factor
+        super().__init__(dim, max_position_embeddings, base, device)
+
+    def _set_cos_sin_cache(self, seq_len, device, dtype):
+        self.max_seq_len_cached = seq_len
+
+        if seq_len > self.max_position_embeddings:
+            base = self.base * (
+                (self.scaling_factor * seq_len / self.max_position_embeddings) - (self.scaling_factor - 1)
+            ) ** (self.dim / (self.dim - 2))
+            inv_freq = 1.0 / (base ** (torch.arange(0, self.dim, 2).float().to(device) / self.dim))
+            self.register_buffer("inv_freq", inv_freq, persistent=False)
+
+        t = torch.arange(self.max_seq_len_cached, device=device, dtype=self.inv_freq.dtype)
+
+        freqs = torch.outer(t, self.inv_freq)
+        # Different from paper, but it uses a different permutation in order to obtain the same calculation
+        emb = torch.cat((freqs, freqs), dim=-1)
+
+        self.register_buffer("cos_cached", emb.cos().to(dtype), persistent=False)
+        self.register_buffer("sin_cached", emb.sin().to(dtype), persistent=False)
+
+
+def rotate_half(x):
+    """Rotates half the hidden dims of the input."""
+    x1 = x[..., : x.shape[-1] // 2]
+    x2 = x[..., x.shape[-1] // 2 :]
+    return torch.cat((-x2, x1), dim=-1)
+
+
+def apply_rotary_pos_emb(q, k, cos, sin, position_ids, unsqueeze_dim=1):
+    """Applies Rotary Position Embedding to the query and key tensors.
+
+    Args:
+        q (`torch.Tensor`): The query tensor.
+        k (`torch.Tensor`): The key tensor.
+        cos (`torch.Tensor`): The cosine part of the rotary embedding.
+        sin (`torch.Tensor`): The sine part of the rotary embedding.
+        position_ids (`torch.Tensor`):
+            The position indices of the tokens corresponding to the query and key tensors. For example, this can be
+            used to pass offsetted position ids when working with a KV-cache.
+        unsqueeze_dim (`int`, *optional*, defaults to 1):
+            The 'unsqueeze_dim' argument specifies the dimension along which to unsqueeze cos[position_ids] and
+            sin[position_ids] so that they can be properly broadcasted to the dimensions of q and k. For example, note
+            that cos[position_ids] and sin[position_ids] have the shape [batch_size, seq_len, head_dim]. Then, if q and
+            k have the shape [batch_size, heads, seq_len, head_dim], then setting unsqueeze_dim=1 makes
+            cos[position_ids] and sin[position_ids] broadcastable to the shapes of q and k. Similarly, if q and k have
+            the shape [batch_size, seq_len, heads, head_dim], then set unsqueeze_dim=2.
+    Returns:
+        `tuple(torch.Tensor)` comprising of the query and key tensors rotated using the Rotary Position Embedding.
+    """
+    # cos = cos[position_ids].unsqueeze(unsqueeze_dim)
+    # sin = sin[position_ids].unsqueeze(unsqueeze_dim)
+    # q_embed = (q * cos) + (rotate_half(q) * sin)
+    # k_embed = (k * cos) + (rotate_half(k) * sin)
+    orig_dtype = k.dtype
+    cos = cos[position_ids].unsqueeze(unsqueeze_dim)  # [bs, 1, seq_len, dim]
+    sin = sin[position_ids].unsqueeze(unsqueeze_dim)  # [bs, 1, seq_len, dim]
+    q_fp32 = q.to(dtype=torch.float32, device=q.device)
+    k_fp32 = k.to(dtype=torch.float32, device=k.device)
+    q_embed = (q_fp32 * cos) + (rotate_half(q_fp32) * sin)
+    k_embed = (k_fp32 * cos) + (rotate_half(k_fp32) * sin)
+    return q_embed.to(dtype=orig_dtype), k_embed.to(dtype=orig_dtype)
+
+class MiniCPMMLP(nn.Module):
+    def __init__(self, config):
+        super().__init__()
+        self.config = config
+        self.hidden_size = config.hidden_size
+        self.intermediate_size = config.intermediate_size
+        self.gate_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.up_proj = nn.Linear(self.hidden_size, self.intermediate_size, bias=False)
+        self.down_proj = nn.Linear(self.intermediate_size, self.hidden_size, bias=False)
+        self.act_fn = ACT2FN[config.hidden_act]
+
+    def forward(self, x):
+        if self.config.pretraining_tp > 1:
+            slice = self.intermediate_size // self.config.pretraining_tp
+            gate_proj_slices = self.gate_proj.weight.split(slice, dim=0)
+            up_proj_slices = self.up_proj.weight.split(slice, dim=0)
+            down_proj_slices = self.down_proj.weight.split(slice, dim=1)
+
+            gate_proj = torch.cat(
+                [F.linear(x, gate_proj_slices[i]) for i in range(self.config.pretraining_tp)], dim=-1
+            )
+            up_proj = torch.cat([F.linear(x, up_proj_slices[i]) for i in range(self.config.pretraining_tp)], dim=-1)
+
+            intermediate_states = (self.act_fn(gate_proj) * up_proj).split(slice, dim=2)
+            down_proj = [
+                F.linear(intermediate_states[i], down_proj_slices[i]) for i in range(self.config.pretraining_tp)
+            ]
+            down_proj = sum(down_proj)
+        else:
+            down_proj = self.down_proj(self.act_fn(self.gate_proj(x)) * self.up_proj(x))
+
+        return down_proj
+
+
+def repeat_kv(hidden_states: torch.Tensor, n_rep: int) -> torch.Tensor:
+    """
+    This is the equivalent of torch.repeat_interleave(x, dim=1, repeats=n_rep). The hidden states go from (batch,
+    num_key_value_heads, seqlen, head_dim) to (batch, num_attention_heads, seqlen, head_dim)
+    """
+    batch, num_key_value_heads, slen, head_dim = hidden_states.shape
+    if n_rep == 1:
+        return hidden_states
+    hidden_states = hidden_states[:, :, None, :, :].expand(batch, num_key_value_heads, n_rep, slen, head_dim)
+    return hidden_states.reshape(batch, num_key_value_heads * n_rep, slen, head_dim)
+
+
+
+class MiniCPMAttention(nn.Module):
+    """Multi-headed attention from 'Attention Is All You Need' paper"""
+
+    def __init__(self, config: MiniCPMConfig, layer_idx: Optional[int] = None):
+        super().__init__()
+        self.config = config
+        self.layer_idx = layer_idx
+        if layer_idx is None:
+            logger.warning_once(
+                f"Instantiating {self.__class__.__name__} without passing `layer_idx` is not recommended and will "
+                "to errors during the forward call, if caching is used. Please make sure to provide a `layer_idx` "
+                "when creating this class."
+            )
+
+        self.attention_dropout = config.attention_dropout
+        self.hidden_size = config.hidden_size
+        self.num_heads = config.num_attention_heads
+        self.head_dim = self.hidden_size // self.num_heads
+        self.num_key_value_heads = config.num_key_value_heads
+        self.num_key_value_groups = self.num_heads // self.num_key_value_heads
+        self.max_position_embeddings = config.max_position_embeddings
+        self.rope_theta = config.rope_theta
+        self.is_causal = True
+
+        if (self.head_dim * self.num_heads) != self.hidden_size:
+            raise ValueError(
+                f"hidden_size must be divisible by num_heads (got `hidden_size`: {self.hidden_size}"
+                f" and `num_heads`: {self.num_heads})."
+            )
+
+        self.q_proj = nn.Linear(self.hidden_size, self.num_heads * self.head_dim, bias=config.attention_bias)
+        self.k_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.v_proj = nn.Linear(self.hidden_size, self.num_key_value_heads * self.head_dim, bias=config.attention_bias)
+        self.o_proj = nn.Linear(self.num_heads * self.head_dim, self.hidden_size, bias=config.attention_bias)
+        self._init_rope()
+
+    def _init_rope(self):
+        if self.config.rope_scaling is None:
+            self.rotary_emb = MiniCPMRotaryEmbedding(
+                self.head_dim,
+                max_position_embeddings=self.max_position_embeddings,
+                base=self.rope_theta,
+            )
+        else:
+            scaling_type = self.config.rope_scaling["type"]
+            scaling_factor = self.config.rope_scaling["factor"]
+            if scaling_type == "linear":
+                self.rotary_emb = MiniCPMLinearScalingRotaryEmbedding(
+                    self.head_dim,
+                    max_position_embeddings=self.max_position_embeddings,
+                    scaling_factor=scaling_factor,
+                    base=self.rope_theta,
+                )
+            elif scaling_type == "dynamic":
+                self.rotary_emb = MiniCPMDynamicNTKScalingRotaryEmbedding(
+                    self.head_dim,
+                    max_position_embeddings=self.max_position_embeddings,
+                    scaling_factor=scaling_factor,
+                    base=self.rope_theta,
+                )
+            else:
+                raise ValueError(f"Unknown RoPE scaling type {scaling_type}")
+
+    def _shape(self, tensor: torch.Tensor, seq_len: int, bsz: int):
+        return tensor.view(bsz, seq_len, self.num_heads, self.head_dim).transpose(1, 2).contiguous()
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        **kwargs,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        if "padding_mask" in kwargs:
+            warnings.warn(
+                "Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
+            )
+
+        bsz, q_len, _ = hidden_states.size()
+
+        if self.config.pretraining_tp > 1:
+            key_value_slicing = (self.num_key_value_heads * self.head_dim) // self.config.pretraining_tp
+            query_slices = self.q_proj.weight.split(
+                (self.num_heads * self.head_dim) // self.config.pretraining_tp, dim=0
+            )
+            key_slices = self.k_proj.weight.split(key_value_slicing, dim=0)
+            value_slices = self.v_proj.weight.split(key_value_slicing, dim=0)
+
+            query_states = [F.linear(hidden_states, query_slices[i]) for i in range(self.config.pretraining_tp)]
+            query_states = torch.cat(query_states, dim=-1)
+
+            key_states = [F.linear(hidden_states, key_slices[i]) for i in range(self.config.pretraining_tp)]
+            key_states = torch.cat(key_states, dim=-1)
+
+            value_states = [F.linear(hidden_states, value_slices[i]) for i in range(self.config.pretraining_tp)]
+            value_states = torch.cat(value_states, dim=-1)
+
+        else:
+            query_states = self.q_proj(hidden_states)
+            key_states = self.k_proj(hidden_states)
+            value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        kv_seq_len = key_states.shape[-2]
+        if past_key_value is not None:
+            if self.layer_idx is None:
+                raise ValueError(
+                    f"The cache structure has changed since version v4.36. If you are using {self.__class__.__name__} "
+                    "for auto-regressive decoding with k/v caching, please make sure to initialize the attention class "
+                    "with a layer index."
+                )
+            kv_seq_len += past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
+        cos, sin = self.rotary_emb(value_states.to(torch.float32), seq_len=kv_seq_len)
+
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
+
+        if past_key_value is not None:
+            cache_kwargs = {"sin": sin, "cos": cos}  # Specific to RoPE models
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        attn_weights = torch.matmul(query_states, key_states.transpose(2, 3)) / math.sqrt(self.head_dim)
+        if attn_weights.size() != (bsz, self.num_heads, q_len, kv_seq_len):
+            raise ValueError(
+                f"Attention weights should be of size {(bsz, self.num_heads, q_len, kv_seq_len)}, but is"
+                f" {attn_weights.size()}"
+            )
+
+        if attention_mask is not None:
+            if attention_mask.size() != (bsz, 1, q_len, kv_seq_len):
+                raise ValueError(
+                    f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)}, but is {attention_mask.size()}"
+                )
+            attn_weights = attn_weights + attention_mask
+
+        # upcast attention to fp32
+        attn_weights = nn.functional.softmax(attn_weights, dim=-1, dtype=torch.float32).to(query_states.dtype)
+        attn_weights = nn.functional.dropout(attn_weights, p=self.attention_dropout, training=self.training)
+        attn_output = torch.matmul(attn_weights, value_states)
+
+        if attn_output.size() != (bsz, self.num_heads, q_len, self.head_dim):
+            raise ValueError(
+                f"`attn_output` should be of size {(bsz, self.num_heads, q_len, self.head_dim)}, but is"
+                f" {attn_output.size()}"
+            )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+
+        attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)
+
+        if self.config.pretraining_tp > 1:
+            attn_output = attn_output.split(self.hidden_size // self.config.pretraining_tp, dim=2)
+            o_proj_slices = self.o_proj.weight.split(self.hidden_size // self.config.pretraining_tp, dim=1)
+            attn_output = sum([F.linear(attn_output[i], o_proj_slices[i]) for i in range(self.config.pretraining_tp)])
+        else:
+            attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+        
+        return attn_output, attn_weights, past_key_value
+
+
+class MiniCPMFlashAttention2(MiniCPMAttention):
+    """
+    MiniCPM flash attention module. This module inherits from `MiniCPMAttention` as the weights of the module stays
+    untouched. The only required change would be on the forward pass where it needs to correctly call the public API of
+    flash attention and deal with padding tokens in case the input contains any of them.
+    """
+
+    def __init__(self, *args, **kwargs):
+        super().__init__(*args, **kwargs)
+
+        # TODO: Should be removed once Flash Attention for RoCm is bumped to 2.1.
+        # flash_attn<2.1 generates top-left aligned causal mask, while what is needed here is bottom-right alignement, that was made default for flash_attn>=2.1. This attribute is used to handle this difference. Reference: https://github.com/Dao-AILab/flash-attention/releases/tag/v2.1.0.
+        # Beware that with flash_attn<2.1, using q_seqlen != k_seqlen (except for the case q_seqlen == 1) produces a wrong mask (top-left).
+        self._flash_attn_uses_top_left_mask = not is_flash_attn_greater_or_equal_2_10()
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.LongTensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+        **kwargs,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        # MiniCPMFlashAttention2 attention does not support output_attentions
+        if "padding_mask" in kwargs:
+            warnings.warn(
+                "Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
+            )
+
+            # overwrite attention_mask with padding_mask
+            attention_mask = kwargs.pop("padding_mask")
+
+        output_attentions = False
+
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        # Flash attention requires the input to have the shape
+        # batch_size x seq_length x head_dim x hidden_dim
+        # therefore we just need to keep the original shape
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        kv_seq_len = key_states.shape[-2]
+        if past_key_value is not None:
+            kv_seq_len += past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
+        cos, sin = self.rotary_emb(value_states.to(torch.float32), seq_len=kv_seq_len)
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
+
+        if past_key_value is not None:
+            cache_kwargs = {"sin": sin, "cos": cos}  # Specific to RoPE models
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        # TODO: These transpose are quite inefficient but Flash Attention requires the layout [batch_size, sequence_length, num_heads, head_dim]. We would need to refactor the KV cache
+        # to be able to avoid many of these transpose/reshape/view.
+        query_states = query_states.transpose(1, 2)
+        key_states = key_states.transpose(1, 2)
+        value_states = value_states.transpose(1, 2)
+
+        dropout_rate = self.attention_dropout if self.training else 0.0
+
+        # In PEFT, usually we cast the layer norms in float32 for training stability reasons
+        # therefore the input hidden states gets silently casted in float32. Hence, we need
+        # cast them back in the correct dtype just to be sure everything works as expected.
+        # This might slowdown training & inference so it is recommended to not cast the LayerNorms
+        # in fp32. (MiniCPMRMSNorm handles it correctly)
+
+        input_dtype = query_states.dtype
+        if input_dtype == torch.float32:
+            # Handle the case where the model is quantized
+            if hasattr(self.config, "_pre_quantization_dtype"):
+                target_dtype = self.config._pre_quantization_dtype
+            else:
+                target_dtype = self.q_proj.weight.dtype
+
+            logger.warning_once(
+                f"The input hidden states seems to be silently casted in float32, this might be related to"
+                f" the fact you have upcasted embedding or layer norm layers in float32. We will cast back the input in"
+                f" {target_dtype}."
+            )
+
+            query_states = query_states.to(target_dtype)
+            key_states = key_states.to(target_dtype)
+            value_states = value_states.to(target_dtype)
+
+        attn_output = self._flash_attention_forward(
+            query_states, key_states, value_states, attention_mask, q_len, dropout=dropout_rate
+        )
+
+        attn_output = attn_output.reshape(bsz, q_len, self.hidden_size).contiguous()
+        attn_output = self.o_proj(attn_output)
+
+        if not output_attentions:
+            attn_weights = None
+
+        return attn_output, attn_weights, past_key_value
+
+    def _flash_attention_forward(
+        self, query_states, key_states, value_states, attention_mask, query_length, dropout=0.0, softmax_scale=None
+    ):
+        """
+        Calls the forward method of Flash Attention - if the input hidden states contain at least one padding token
+        first unpad the input, then computes the attention scores and pad the final attention scores.
+
+        Args:
+            query_states (`torch.Tensor`):
+                Input query states to be passed to Flash Attention API
+            key_states (`torch.Tensor`):
+                Input key states to be passed to Flash Attention API
+            value_states (`torch.Tensor`):
+                Input value states to be passed to Flash Attention API
+            attention_mask (`torch.Tensor`):
+                The padding mask - corresponds to a tensor of size `(batch_size, seq_len)` where 0 stands for the
+                position of padding tokens and 1 for the position of non-padding tokens.
+            dropout (`int`, *optional*):
+                Attention dropout
+            softmax_scale (`float`, *optional*):
+                The scaling of QK^T before applying softmax. Default to 1 / sqrt(head_dim)
+        """
+        if not self._flash_attn_uses_top_left_mask:
+            causal = self.is_causal
+        else:
+            # TODO: Remove the `query_length != 1` check once Flash Attention for RoCm is bumped to 2.1. For details, please see the comment in MiniCPMFlashAttention2 __init__.
+            causal = self.is_causal and query_length != 1
+        # Contains at least one padding token in the sequence
+        if attention_mask is not None:
+            batch_size = query_states.shape[0]
+            query_states, key_states, value_states, indices_q, cu_seq_lens, max_seq_lens = self._upad_input(
+                query_states, key_states, value_states, attention_mask, query_length
+            )
+
+            cu_seqlens_q, cu_seqlens_k = cu_seq_lens
+            max_seqlen_in_batch_q, max_seqlen_in_batch_k = max_seq_lens
+            attn_output_unpad = flash_attn_varlen_func(
+                query_states,
+                key_states,
+                value_states,
+                cu_seqlens_q=cu_seqlens_q,
+                cu_seqlens_k=cu_seqlens_k,
+                max_seqlen_q=max_seqlen_in_batch_q,
+                max_seqlen_k=max_seqlen_in_batch_k,
+                dropout_p=dropout,
+                softmax_scale=softmax_scale,
+                causal=causal,
+            )
+
+            attn_output = pad_input(attn_output_unpad, indices_q, batch_size, query_length)
+        else:
+            attn_output = flash_attn_func(
+                query_states, key_states, value_states, dropout, softmax_scale=softmax_scale, causal=causal
+            )
+
+        return attn_output
+
+    def _upad_input(self, query_layer, key_layer, value_layer, attention_mask, query_length):
+        indices_k, cu_seqlens_k, max_seqlen_in_batch_k = _get_unpad_data(attention_mask)
+        batch_size, kv_seq_len, num_key_value_heads, head_dim = key_layer.shape
+
+        key_layer = index_first_axis(
+            key_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
+        )
+        value_layer = index_first_axis(
+            value_layer.reshape(batch_size * kv_seq_len, num_key_value_heads, head_dim), indices_k
+        )
+        if query_length == kv_seq_len:
+            query_layer = index_first_axis(
+                query_layer.reshape(batch_size * kv_seq_len, self.num_heads, head_dim), indices_k
+            )
+            cu_seqlens_q = cu_seqlens_k
+            max_seqlen_in_batch_q = max_seqlen_in_batch_k
+            indices_q = indices_k
+        elif query_length == 1:
+            max_seqlen_in_batch_q = 1
+            cu_seqlens_q = torch.arange(
+                batch_size + 1, dtype=torch.int32, device=query_layer.device
+            )  # There is a memcpy here, that is very bad.
+            indices_q = cu_seqlens_q[:-1]
+            query_layer = query_layer.squeeze(1)
+        else:
+            # The -q_len: slice assumes left padding.
+            attention_mask = attention_mask[:, -query_length:]
+            query_layer, indices_q, cu_seqlens_q, max_seqlen_in_batch_q = unpad_input(query_layer, attention_mask)
+
+        return (
+            query_layer,
+            key_layer,
+            value_layer,
+            indices_q,
+            (cu_seqlens_q, cu_seqlens_k),
+            (max_seqlen_in_batch_q, max_seqlen_in_batch_k),
+        )
+
+
+class MiniCPMSdpaAttention(MiniCPMAttention):
+    """
+    MiniCPM attention module using torch.nn.functional.scaled_dot_product_attention. This module inherits from
+    `MiniCPMAttention` as the weights of the module stays untouched. The only changes are on the forward pass to adapt to
+    SDPA API.
+    """
+
+    # Adapted from MiniCPMAttention.forward
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Cache] = None,
+        output_attentions: bool = False,
+        use_cache: bool = False,
+    ) -> Tuple[torch.Tensor, Optional[torch.Tensor], Optional[Tuple[torch.Tensor]]]:
+        if output_attentions:
+            # TODO: Improve this warning with e.g. `model.config.attn_implementation = "manual"` once this is implemented.
+            logger.warning_once(
+                "MiniCPMModel is using MiniCPMSdpaAttention, but `torch.nn.functional.scaled_dot_product_attention` does not support `output_attentions=True`. Falling back to the manual attention implementation, "
+                'but specifying the manual implementation will be required from Transformers version v5.0.0 onwards. This warning can be removed using the argument `attn_implementation="eager"` when loading the model.'
+            )
+            return super().forward(
+                hidden_states=hidden_states,
+                attention_mask=attention_mask,
+                position_ids=position_ids,
+                past_key_value=past_key_value,
+                output_attentions=output_attentions,
+                use_cache=use_cache,
+            )
+
+        bsz, q_len, _ = hidden_states.size()
+
+        query_states = self.q_proj(hidden_states)
+        key_states = self.k_proj(hidden_states)
+        value_states = self.v_proj(hidden_states)
+
+        query_states = query_states.view(bsz, q_len, self.num_heads, self.head_dim).transpose(1, 2)
+        key_states = key_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+        value_states = value_states.view(bsz, q_len, self.num_key_value_heads, self.head_dim).transpose(1, 2)
+
+        kv_seq_len = key_states.shape[-2]
+        if past_key_value is not None:
+            kv_seq_len += past_key_value.get_usable_length(kv_seq_len, self.layer_idx)
+        cos, sin = self.rotary_emb(value_states, seq_len=kv_seq_len)
+
+        query_states, key_states = apply_rotary_pos_emb(query_states, key_states, cos, sin, position_ids)
+
+        if past_key_value is not None:
+            cache_kwargs = {"sin": sin, "cos": cos}  # Specific to RoPE models
+            key_states, value_states = past_key_value.update(key_states, value_states, self.layer_idx, cache_kwargs)
+
+        key_states = repeat_kv(key_states, self.num_key_value_groups)
+        value_states = repeat_kv(value_states, self.num_key_value_groups)
+
+        if attention_mask is not None:
+            if attention_mask.size() != (bsz, 1, q_len, kv_seq_len):
+                raise ValueError(
+                    f"Attention mask should be of size {(bsz, 1, q_len, kv_seq_len)}, but is {attention_mask.size()}"
+                )
+
+        # SDPA with memory-efficient backend is currently (torch==2.1.2) bugged with non-contiguous inputs with custom attn_mask,
+        # Reference: https://github.com/pytorch/pytorch/issues/112577.
+        if query_states.device.type == "cuda" and attention_mask is not None:
+            query_states = query_states.contiguous()
+            key_states = key_states.contiguous()
+            value_states = value_states.contiguous()
+
+        attn_output = torch.nn.functional.scaled_dot_product_attention(
+            query_states,
+            key_states,
+            value_states,
+            attn_mask=attention_mask,
+            dropout_p=self.attention_dropout if self.training else 0.0,
+            # The q_len > 1 is necessary to match with AttentionMaskConverter.to_causal_4d that does not create a causal mask in case q_len == 1.
+            is_causal=self.is_causal and attention_mask is None and q_len > 1,
+        )
+
+        attn_output = attn_output.transpose(1, 2).contiguous()
+        attn_output = attn_output.reshape(bsz, q_len, self.hidden_size)
+
+        attn_output = self.o_proj(attn_output)
+
+        return attn_output, None, past_key_value
+
+
+MINICPM_ATTENTION_CLASSES = {
+    "eager": MiniCPMAttention,
+    "flash_attention_2": MiniCPMFlashAttention2,
+    "sdpa": MiniCPMSdpaAttention,
+}
+
+
+class MiniCPMDecoderLayer(nn.Module):
+    def __init__(self, config: MiniCPMConfig, layer_idx: int):
+        super().__init__()
+        self.hidden_size = config.hidden_size
+        self.self_attn = MINICPM_ATTENTION_CLASSES[config._attn_implementation](config=config, layer_idx=layer_idx)
+
+        self.mlp = MiniCPMMLP(config)
+        self.input_layernorm = MiniCPMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+        self.post_attention_layernorm = MiniCPMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+        self.scale_depth = config.scale_depth
+        self.num_hidden_layers = config.num_hidden_layers
+
+    def forward(
+        self,
+        hidden_states: torch.Tensor,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_value: Optional[Tuple[torch.Tensor]] = None,
+        output_attentions: Optional[bool] = False,
+        use_cache: Optional[bool] = False,
+        **kwargs,
+    ) -> Tuple[torch.FloatTensor, Optional[Tuple[torch.FloatTensor, torch.FloatTensor]]]:
+        """
+        Args:
+            hidden_states (`torch.FloatTensor`): input to the layer of shape `(batch, seq_len, embed_dim)`
+            attention_mask (`torch.FloatTensor`, *optional*):
+                attention mask of size `(batch_size, sequence_length)` if flash attention is used or `(batch_size, 1,
+                query_sequence_length, key_sequence_length)` if default attention is used.
+            output_attentions (`bool`, *optional*):
+                Whether or not to return the attentions tensors of all attention layers. See `attentions` under
+                returned tensors for more detail.
+            use_cache (`bool`, *optional*):
+                If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding
+                (see `past_key_values`).
+            past_key_value (`Tuple(torch.FloatTensor)`, *optional*): cached past key and value projection states
+        """
+        if "padding_mask" in kwargs:
+            warnings.warn(
+                "Passing `padding_mask` is deprecated and will be removed in v4.37. Please make sure use `attention_mask` instead.`"
+            )
+
+        residual = hidden_states
+        hidden_states = self.input_layernorm(hidden_states)
+        # Self Attention
+        hidden_states, self_attn_weights, present_key_value = self.self_attn(
+            hidden_states=hidden_states,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_value=past_key_value,
+            output_attentions=output_attentions,
+            use_cache=use_cache,
+            **kwargs,
+        )
+        
+        hidden_states = residual + hidden_states * (self.scale_depth / math.sqrt(self.num_hidden_layers))
+
+        # Fully Connected
+        residual = hidden_states
+        hidden_states = self.post_attention_layernorm(hidden_states)
+
+        hidden_states = self.mlp(hidden_states)
+        hidden_states = residual + hidden_states * (self.scale_depth / math.sqrt(self.num_hidden_layers))
+
+        outputs = (hidden_states,)
+
+        if output_attentions:
+            outputs += (self_attn_weights,)
+
+        if use_cache:
+            outputs += (present_key_value,)
+
+        return outputs
+
+
+MINICPM_START_DOCSTRING = r"""
+    This model inherits from [`PreTrainedModel`]. Check the superclass documentation for the generic methods the
+    library implements for all its model (such as downloading or saving, resizing the input embeddings, pruning heads
+    etc.)
+
+    This model is also a PyTorch [torch.nn.Module](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) subclass.
+    Use it as a regular PyTorch Module and refer to the PyTorch documentation for all matter related to general usage
+    and behavior.
+
+    Parameters:
+        config ([`MiniCPMConfig`]):
+            Model configuration class with all the parameters of the model. Initializing with a config file does not
+            load the weights associated with the model, only the configuration. Check out the
+            [`~PreTrainedModel.from_pretrained`] method to load the model weights.
+"""
+
+
+@add_start_docstrings(
+    "The bare MiniCPM Model outputting raw hidden-states without any specific head on top.",
+    MINICPM_START_DOCSTRING,
+)
+class MiniCPMPreTrainedModel(PreTrainedModel):
+    config_class = MiniCPMConfig
+    base_model_prefix = "model"
+    supports_gradient_checkpointing = True
+    _no_split_modules = ["MiniCPMDecoderLayer"]
+    _skip_keys_device_placement = "past_key_values"
+    _supports_flash_attn_2 = True
+    _supports_sdpa = True
+    _supports_cache_class = True
+
+    def _init_weights(self, module):
+        std = self.config.initializer_range
+        if isinstance(module, nn.Linear):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.bias is not None:
+                module.bias.data.zero_()
+        elif isinstance(module, nn.Embedding):
+            module.weight.data.normal_(mean=0.0, std=std)
+            if module.padding_idx is not None:
+                module.weight.data[module.padding_idx].zero_()
+
+
+MINICPM_INPUTS_DOCSTRING = r"""
+    Args:
+        input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
+            Indices of input sequence tokens in the vocabulary. Padding will be ignored by default should you provide
+            it.
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            [What are input IDs?](../glossary#input-ids)
+        attention_mask (`torch.Tensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Mask to avoid performing attention on padding token indices. Mask values selected in `[0, 1]`:
+
+            - 1 for tokens that are **not masked**,
+            - 0 for tokens that are **masked**.
+
+            [What are attention masks?](../glossary#attention-mask)
+
+            Indices can be obtained using [`AutoTokenizer`]. See [`PreTrainedTokenizer.encode`] and
+            [`PreTrainedTokenizer.__call__`] for details.
+
+            If `past_key_values` is used, optionally only the last `input_ids` have to be input (see
+            `past_key_values`).
+
+            If you want to change padding behavior, you should read [`modeling_opt._prepare_decoder_attention_mask`]
+            and modify to your needs. See diagram 1 in [the paper](https://arxiv.org/abs/1910.13461) for more
+            information on the default strategy.
+
+            - 1 indicates the head is **not masked**,
+            - 0 indicates the head is **masked**.
+        position_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+            Indices of positions of each input sequence tokens in the position embeddings. Selected in the range `[0,
+            config.n_positions - 1]`.
+
+            [What are position IDs?](../glossary#position-ids)
+        past_key_values (`Cache` or `tuple(tuple(torch.FloatTensor))`, *optional*):
+            Pre-computed hidden-states (key and values in the self-attention blocks and in the cross-attention
+            blocks) that can be used to speed up sequential decoding. This typically consists in the `past_key_values`
+            returned by the model at a previous stage of decoding, when `use_cache=True` or `config.use_cache=True`.
+
+            Two formats are allowed:
+            - a [`~cache_utils.Cache`] instance;
+            - Tuple of `tuple(torch.FloatTensor)` of length `config.n_layers`, with each tuple having 2 tensors of
+            shape `(batch_size, num_heads, sequence_length, embed_size_per_head)`). This is also known as the legacy
+            cache format.
+
+            The model will output the same cache format that is fed as input. If no `past_key_values` are passed, the
+            legacy cache format will be returned.
+
+            If `past_key_values` are used, the user can optionally input only the last `input_ids` (those that don't
+            have their past key value states given to this model) of shape `(batch_size, 1)` instead of all `input_ids`
+            of shape `(batch_size, sequence_length)`.
+        inputs_embeds (`torch.FloatTensor` of shape `(batch_size, sequence_length, hidden_size)`, *optional*):
+            Optionally, instead of passing `input_ids` you can choose to directly pass an embedded representation. This
+            is useful if you want more control over how to convert `input_ids` indices into associated vectors than the
+            model's internal embedding lookup matrix.
+        use_cache (`bool`, *optional*):
+            If set to `True`, `past_key_values` key value states are returned and can be used to speed up decoding (see
+            `past_key_values`).
+        output_attentions (`bool`, *optional*):
+            Whether or not to return the attentions tensors of all attention layers. See `attentions` under returned
+            tensors for more detail.
+        output_hidden_states (`bool`, *optional*):
+            Whether or not to return the hidden states of all layers. See `hidden_states` under returned tensors for
+            more detail.
+        return_dict (`bool`, *optional*):
+            Whether or not to return a [`~utils.ModelOutput`] instead of a plain tuple.
+"""
+
+
+@add_start_docstrings(
+    "The bare MiniCPM Model outputting raw hidden-states without any specific head on top.",
+    MINICPM_START_DOCSTRING,
+)
+class MiniCPMModel(MiniCPMPreTrainedModel):
+    """
+    Transformer decoder consisting of *config.num_hidden_layers* layers. Each layer is a [`MiniCPMDecoderLayer`]
+
+    Args:
+        config: MiniCPMConfig
+    """
+
+    def __init__(self, config: MiniCPMConfig):
+        super().__init__(config)
+        self.padding_idx = config.pad_token_id
+        self.vocab_size = config.vocab_size
+
+        self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.padding_idx)
+        self.layers = nn.ModuleList(
+            [MiniCPMDecoderLayer(config, layer_idx) for layer_idx in range(config.num_hidden_layers)]
+        )
+        self._use_sdpa = config._attn_implementation == "sdpa"
+        self._use_flash_attention_2 = config._attn_implementation == "flash_attention_2"
+
+        self.norm = MiniCPMRMSNorm(config.hidden_size, eps=config.rms_norm_eps)
+
+        self.gradient_checkpointing = False
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.embed_tokens = value
+
+    @add_start_docstrings_to_model_forward(MINICPM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, BaseModelOutputWithPast]:
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        use_cache = use_cache if use_cache is not None else self.config.use_cache
+
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # retrieve input_ids and inputs_embeds
+        if input_ids is not None and inputs_embeds is not None:
+            raise ValueError("You cannot specify both input_ids and inputs_embeds at the same time")
+        elif input_ids is not None:
+            batch_size, seq_length = input_ids.shape[:2]
+        elif inputs_embeds is not None:
+            batch_size, seq_length = inputs_embeds.shape[:2]
+        else:
+            raise ValueError("You have to specify either input_ids or inputs_embeds")
+
+        if self.gradient_checkpointing and self.training:
+            if use_cache:
+                logger.warning_once(
+                    "`use_cache=True` is incompatible with gradient checkpointing. Setting `use_cache=False`..."
+                )
+                use_cache = False
+
+        past_key_values_length = 0
+        if use_cache:
+            use_legacy_cache = not isinstance(past_key_values, Cache)
+            if use_legacy_cache:
+                past_key_values = DynamicCache.from_legacy_cache(past_key_values)
+            past_key_values_length = past_key_values.get_usable_length(seq_length)
+
+        if position_ids is None:
+            device = input_ids.device if input_ids is not None else inputs_embeds.device
+            position_ids = torch.arange(
+                past_key_values_length, seq_length + past_key_values_length, dtype=torch.long, device=device
+            )
+            position_ids = position_ids.unsqueeze(0)
+
+        if inputs_embeds is None:
+            inputs_embeds = self.embed_tokens(input_ids) * self.config.scale_emb
+
+        if self._use_flash_attention_2:
+            # 2d mask is passed through the layers
+            attention_mask = attention_mask if (attention_mask is not None and 0 in attention_mask) else None
+        elif self._use_sdpa and not output_attentions:
+            # output_attentions=True can not be supported when using SDPA, and we fall back on
+            # the manual implementation that requires a 4D causal mask in all cases.
+            attention_mask = _prepare_4d_causal_attention_mask_for_sdpa(
+                attention_mask,
+                (batch_size, seq_length),
+                inputs_embeds,
+                past_key_values_length,
+            )
+        else:
+            # 4d mask is passed through the layers
+            attention_mask = _prepare_4d_causal_attention_mask(
+                attention_mask, (batch_size, seq_length), inputs_embeds, past_key_values_length
+            )
+
+        # embed positions
+        hidden_states = inputs_embeds
+
+        # decoder layers
+        all_hidden_states = () if output_hidden_states else None
+        all_self_attns = () if output_attentions else None
+        next_decoder_cache = None
+
+        for decoder_layer in self.layers:
+            if output_hidden_states:
+                all_hidden_states += (hidden_states,)
+
+            if self.gradient_checkpointing and self.training:
+                layer_outputs = self._gradient_checkpointing_func(
+                    decoder_layer.__call__,
+                    hidden_states,
+                    attention_mask,
+                    position_ids,
+                    past_key_values,
+                    output_attentions,
+                    use_cache,
+                )
+            else:
+                layer_outputs = decoder_layer(
+                    hidden_states,
+                    attention_mask=attention_mask,
+                    position_ids=position_ids,
+                    past_key_value=past_key_values,
+                    output_attentions=output_attentions,
+                    use_cache=use_cache,
+                )
+
+            hidden_states = layer_outputs[0]
+
+            if use_cache:
+                next_decoder_cache = layer_outputs[2 if output_attentions else 1]
+
+            if output_attentions:
+                all_self_attns += (layer_outputs[1],)
+
+        hidden_states = self.norm(hidden_states)
+
+        # add hidden states from the last decoder layer
+        if output_hidden_states:
+            all_hidden_states += (hidden_states,)
+
+        next_cache = None
+        if use_cache:
+            next_cache = next_decoder_cache.to_legacy_cache() if use_legacy_cache else next_decoder_cache
+        if not return_dict:
+            return tuple(v for v in [hidden_states, next_cache, all_hidden_states, all_self_attns] if v is not None)
+        return BaseModelOutputWithPast(
+            last_hidden_state=hidden_states,
+            past_key_values=next_cache,
+            hidden_states=all_hidden_states,
+            attentions=all_self_attns,
+        )
+
+
+class MiniCPMForCausalLM(MiniCPMPreTrainedModel):
+    _tied_weights_keys = ["lm_head.weight"]
+
+    def __init__(self, config):
+        super().__init__(config)
+        self.model = MiniCPMModel(config)
+        self.vocab_size = config.vocab_size
+        self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    def get_output_embeddings(self):
+        return self.lm_head
+
+    def set_output_embeddings(self, new_embeddings):
+        self.lm_head = new_embeddings
+
+    def set_decoder(self, decoder):
+        self.model = decoder
+
+    def get_decoder(self):
+        return self.model
+
+    @add_start_docstrings_to_model_forward(MINICPM_INPUTS_DOCSTRING)
+    @replace_return_docstrings(output_type=CausalLMOutputWithPast, config_class=_CONFIG_FOR_DOC)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, CausalLMOutputWithPast]:
+        r"""
+        Args:
+            labels (`torch.LongTensor` of shape `(batch_size, sequence_length)`, *optional*):
+                Labels for computing the masked language modeling loss. Indices should either be in `[0, ...,
+                config.vocab_size]` or -100 (see `input_ids` docstring). Tokens with indices set to `-100` are ignored
+                (masked), the loss is only computed for the tokens with labels in `[0, ..., config.vocab_size]`.
+
+        Returns:
+
+        Example:
+
+        ```python
+        >>> from transformers import AutoTokenizer, MiniCPMForCausalLM
+
+        >>> model = MiniCPMForCausalLM.from_pretrained(PATH_TO_CONVERTED_WEIGHTS)
+        >>> tokenizer = AutoTokenizer.from_pretrained(PATH_TO_CONVERTED_TOKENIZER)
+
+        >>> prompt = "Hey, are you conscious? Can you talk to me?"
+        >>> inputs = tokenizer(prompt, return_tensors="pt")
+
+        >>> # Generate
+        >>> generate_ids = model.generate(inputs.input_ids, max_length=30)
+        >>> tokenizer.batch_decode(generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False)[0]
+        "Hey, are you conscious? Can you talk to me?\nI'm not conscious, but I can talk to you."
+        ```"""
+        output_attentions = output_attentions if output_attentions is not None else self.config.output_attentions
+        output_hidden_states = (
+            output_hidden_states if output_hidden_states is not None else self.config.output_hidden_states
+        )
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        # decoder outputs consists of (dec_features, layer_state, dec_hidden, dec_attn)
+        outputs = self.model(
+            input_ids=input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+
+        hidden_states = outputs[0]
+        if self.config.pretraining_tp > 1:
+            lm_head_slices = self.lm_head.weight.split(self.vocab_size // self.config.pretraining_tp, dim=0)
+            logits = [F.linear(hidden_states, lm_head_slices[i]) for i in range(self.config.pretraining_tp)]
+            logits = torch.cat(logits, dim=-1)
+        else:
+            logits = self.lm_head(hidden_states / (self.config.hidden_size / self.config.dim_model_base))
+        logits = logits.float()
+
+        loss = None
+        if labels is not None:
+            # Shift so that tokens < n predict n
+            shift_logits = logits[..., :-1, :].contiguous()
+            shift_labels = labels[..., 1:].contiguous()
+            # Flatten the tokens
+            loss_fct = CrossEntropyLoss()
+            shift_logits = shift_logits.view(-1, self.config.vocab_size)
+            shift_labels = shift_labels.view(-1)
+            # Enable model parallelism
+            shift_labels = shift_labels.to(shift_logits.device)
+            loss = loss_fct(shift_logits, shift_labels)
+
+        if not return_dict:
+            output = (logits,) + outputs[1:]
+            return (loss,) + output if loss is not None else output
+
+        return CausalLMOutputWithPast(
+            loss=loss,
+            logits=logits,
+            past_key_values=outputs.past_key_values,
+            hidden_states=outputs.hidden_states,
+            attentions=outputs.attentions,
+        )
+
+    def prepare_inputs_for_generation(
+        self, input_ids, past_key_values=None, attention_mask=None, inputs_embeds=None, **kwargs
+    ):
+        if past_key_values is not None:
+            if isinstance(past_key_values, Cache):
+                cache_length = past_key_values.get_seq_length()
+                past_length = past_key_values.seen_tokens
+                max_cache_length = past_key_values.get_max_cache_shape()
+            else:
+                cache_length = past_length = past_key_values[0][0].shape[2]
+                max_cache_length = None
+
+            # Keep only the unprocessed tokens:
+            # 1 - If the length of the attention_mask exceeds the length of input_ids, then we are in a setting where
+            # some of the inputs are exclusivelly passed as part of the cache (e.g. when passing input_embeds as
+            # input)
+            if attention_mask is not None and attention_mask.shape[1] > input_ids.shape[1]:
+                input_ids = input_ids[:, -(attention_mask.shape[1] - past_length) :]
+            # 2 - If the past_length is smaller than input_ids', then input_ids holds all input tokens. We can discard
+            # input_ids based on the past_length.
+            elif past_length < input_ids.shape[1]:
+                input_ids = input_ids[:, past_length:]
+            # 3 - Otherwise (past_length >= input_ids.shape[1]), let's assume input_ids only has unprocessed tokens.
+
+            # If we are about to go beyond the maximum cache length, we need to crop the input attention mask.
+            if (
+                max_cache_length is not None
+                and attention_mask is not None
+                and cache_length + input_ids.shape[1] > max_cache_length
+            ):
+                attention_mask = attention_mask[:, -max_cache_length:]
+
+        position_ids = kwargs.get("position_ids", None)
+        if attention_mask is not None and position_ids is None:
+            # create position_ids on the fly for batch generation
+            position_ids = attention_mask.long().cumsum(-1) - 1
+            position_ids.masked_fill_(attention_mask == 0, 1)
+            if past_key_values:
+                position_ids = position_ids[:, -input_ids.shape[1] :]
+
+        # if `inputs_embeds` are passed, we only want to use them in the 1st generation step
+        if inputs_embeds is not None and past_key_values is None:
+            model_inputs = {"inputs_embeds": inputs_embeds}
+        else:
+            model_inputs = {"input_ids": input_ids}
+
+        model_inputs.update(
+            {
+                "position_ids": position_ids,
+                "past_key_values": past_key_values,
+                "use_cache": kwargs.get("use_cache"),
+                "attention_mask": attention_mask,
+            }
+        )
+        return model_inputs
+
+    @staticmethod
+    def _reorder_cache(past_key_values, beam_idx):
+        reordered_past = ()
+        for layer_past in past_key_values:
+            reordered_past += (
+                tuple(past_state.index_select(0, beam_idx.to(past_state.device)) for past_state in layer_past),
+            )
+        return reordered_past
+    
+    @torch.inference_mode()
+    def chat(self, tokenizer, query: str, history: List[Dict] = None, role: str = "user",
+             max_length: int = 4096, num_beams=1, do_sample=True, top_p=0.8, temperature=0.3, logits_processor=None,
+             **kwargs):
+        if history is None:
+            history = []
+        if logits_processor:
+            gen_kwargs = {"max_new_tokens": max_length, "num_beams": num_beams, "do_sample": do_sample, "top_p": top_p,
+                        "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        else:
+            gen_kwargs = {"max_new_tokens": max_length, "num_beams": num_beams, "do_sample": do_sample, "top_p": top_p,
+                        "temperature": temperature, "logits_processor": logits_processor, **kwargs}
+        
+        history.append({"role": role, "content": query})
+        history_str = tokenizer.apply_chat_template(history, tokenize=False, add_generation_prompt=False)
+        inputs = tokenizer(history_str, return_tensors='pt').to(self.device)
+        outputs = self.generate(**inputs, **gen_kwargs)
+        outputs = outputs.tolist()[0][len(inputs["input_ids"][0]):-1]
+        response = tokenizer.decode(outputs)
+        pattern = re.compile(r".*?(?=<AI>|<用户>)", re.DOTALL)
+        matches = pattern.findall(response)
+        if len(matches) > 0:
+            response = matches[0]
+        history.append({"role": "assistant", "content": response})
+        return response, history
+
+
+@add_start_docstrings(
+    """
+    The MiniCPM Model transformer with a sequence classification head on top (linear layer).
+
+    [`MiniCPMForSequenceClassification`] uses the last token in order to do the classification, as other causal models
+    (e.g. GPT-2) do.
+
+    Since it does classification on the last token, it requires to know the position of the last token. If a
+    `pad_token_id` is defined in the configuration, it finds the last token that is not a padding token in each row. If
+    no `pad_token_id` is defined, it simply takes the last value in each row of the batch. Since it cannot guess the
+    padding tokens when `inputs_embeds` are passed instead of `input_ids`, it does the same (take the last value in
+    each row of the batch).
+    """,
+    MINICPM_START_DOCSTRING,
+)
+class MiniCPMForSequenceClassification(MiniCPMPreTrainedModel):
+    def __init__(self, config):
+        super().__init__(config)
+        self.num_labels = config.num_labels
+        self.model = MiniCPMModel(config)
+        self.score = nn.Linear(config.hidden_size, self.num_labels, bias=False)
+
+        # Initialize weights and apply final processing
+        self.post_init()
+
+    def get_input_embeddings(self):
+        return self.model.embed_tokens
+
+    def set_input_embeddings(self, value):
+        self.model.embed_tokens = value
+
+    @add_start_docstrings_to_model_forward(MINICPM_INPUTS_DOCSTRING)
+    def forward(
+        self,
+        input_ids: torch.LongTensor = None,
+        attention_mask: Optional[torch.Tensor] = None,
+        position_ids: Optional[torch.LongTensor] = None,
+        past_key_values: Optional[List[torch.FloatTensor]] = None,
+        inputs_embeds: Optional[torch.FloatTensor] = None,
+        labels: Optional[torch.LongTensor] = None,
+        use_cache: Optional[bool] = None,
+        output_attentions: Optional[bool] = None,
+        output_hidden_states: Optional[bool] = None,
+        return_dict: Optional[bool] = None,
+    ) -> Union[Tuple, SequenceClassifierOutputWithPast]:
+        r"""
+        labels (`torch.LongTensor` of shape `(batch_size,)`, *optional*):
+            Labels for computing the sequence classification/regression loss. Indices should be in `[0, ...,
+            config.num_labels - 1]`. If `config.num_labels == 1` a regression loss is computed (Mean-Square loss), If
+            `config.num_labels > 1` a classification loss is computed (Cross-Entropy).
+        """
+        return_dict = return_dict if return_dict is not None else self.config.use_return_dict
+
+        transformer_outputs = self.model(
+            input_ids,
+            attention_mask=attention_mask,
+            position_ids=position_ids,
+            past_key_values=past_key_values,
+            inputs_embeds=inputs_embeds,
+            use_cache=use_cache,
+            output_attentions=output_attentions,
+            output_hidden_states=output_hidden_states,
+            return_dict=return_dict,
+        )
+        hidden_states = transformer_outputs[0]
+        logits = self.score(hidden_states)
+
+        if input_ids is not None:
+            batch_size = input_ids.shape[0]
+        else:
+            batch_size = inputs_embeds.shape[0]
+
+        if self.config.pad_token_id is None and batch_size != 1:
+            raise ValueError("Cannot handle batch sizes > 1 if no padding token is defined.")
+        if self.config.pad_token_id is None:
+            sequence_lengths = -1
+        else:
+            if input_ids is not None:
+                sequence_lengths = (torch.eq(input_ids, self.config.pad_token_id).int().argmax(-1) - 1).to(
+                    logits.device
+                )
+            else:
+                sequence_lengths = -1
+
+        pooled_logits = logits[torch.arange(batch_size, device=logits.device), sequence_lengths]
+
+        loss = None
+        if labels is not None:
+            labels = labels.to(logits.device)
+            if self.config.problem_type is None:
+                if self.num_labels == 1:
+                    self.config.problem_type = "regression"
+                elif self.num_labels > 1 and (labels.dtype == torch.long or labels.dtype == torch.int):
+                    self.config.problem_type = "single_label_classification"
+                else:
+                    self.config.problem_type = "multi_label_classification"
+
+            if self.config.problem_type == "regression":
+                loss_fct = MSELoss()
+                if self.num_labels == 1:
+                    loss = loss_fct(pooled_logits.squeeze(), labels.squeeze())
+                else:
+                    loss = loss_fct(pooled_logits, labels)
+            elif self.config.problem_type == "single_label_classification":
+                loss_fct = CrossEntropyLoss()
+                loss = loss_fct(pooled_logits.view(-1, self.num_labels), labels.view(-1))
+            elif self.config.problem_type == "multi_label_classification":
+                loss_fct = BCEWithLogitsLoss()
+                loss = loss_fct(pooled_logits, labels)
+        if not return_dict:
+            output = (pooled_logits,) + transformer_outputs[1:]
+            return ((loss,) + output) if loss is not None else output
+
+        return SequenceClassifierOutputWithPast(
+            loss=loss,
+            logits=pooled_logits,
+            past_key_values=transformer_outputs.past_key_values,
+            hidden_states=transformer_outputs.hidden_states,
+            attentions=transformer_outputs.attentions,
+        )

runs/minicpm-2b-sft-bf16/events.out.tfevents.1748344962.n176-050-036

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b032aeeb217decfbc4cee1e4cd93f53d6973eae107faf87001ccf146394cd983
+size 214748

runs/minicpm-2b-sft-bf16/events.out.tfevents.1748349034.n176-050-036

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaf22c1ac017f2718d2d83fe4290a2bb2d1d169fec6db8e60978a86e28a74fc0
+size 311

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ea94864e21236075fe2eb4b324d2626b2e3e9e5135b67b3465b7ad88da63f6b
+size 11177403

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c9aafcd7da1f5611dab6be545db74d5552a2ccc9c2a12c72ea7be63aac4a25d7
+size 1994871

tokenizer_config.json

@@ -0,0 +1,45 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 35.0,
+    "total_flos": 1.12061156850611e+18,
+    "train_loss": 0.2908882966189056,
+    "train_runtime": 4060.664,
+    "train_samples_per_second": 119.463,
+    "train_steps_per_second": 0.241
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c88b2c1464e331d6bf50905b5d21ad9325669e5991e295ad9f212aa0bb68f9c
+size 7697