Model save
Browse files- README.md +68 -0
- all_results.json +9 -0
- config.json +30 -0
- generation_config.json +7 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- runs/Oct30_23-27-03_ip-26-0-164-18/events.out.tfevents.1730332615.ip-26-0-164-18.40006.0 +3 -0
- special_tokens_map.json +28 -0
- tokenizer.json +0 -0
- tokenizer_config.json +154 -0
- train_results.json +9 -0
- trainer_state.json +2564 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: HuggingFaceTB/SmolLM2-1.7B-8k
|
| 3 |
+
tags:
|
| 4 |
+
- trl
|
| 5 |
+
- sft
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
datasets:
|
| 8 |
+
- generator
|
| 9 |
+
model-index:
|
| 10 |
+
- name: smollm2-1.7B-8k-mix7-ep2-v2
|
| 11 |
+
results: []
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 15 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 16 |
+
|
| 17 |
+
[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/loubnabnl/huggingface/runs/6rp7tpcv)
|
| 18 |
+
# smollm2-1.7B-8k-mix7-ep2-v2
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-1.7B-8k](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-8k) on the generator dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 1.0630
|
| 23 |
+
|
| 24 |
+
## Model description
|
| 25 |
+
|
| 26 |
+
More information needed
|
| 27 |
+
|
| 28 |
+
## Intended uses & limitations
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Training and evaluation data
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training procedure
|
| 37 |
+
|
| 38 |
+
### Training hyperparameters
|
| 39 |
+
|
| 40 |
+
The following hyperparameters were used during training:
|
| 41 |
+
- learning_rate: 0.0003
|
| 42 |
+
- train_batch_size: 4
|
| 43 |
+
- eval_batch_size: 4
|
| 44 |
+
- seed: 42
|
| 45 |
+
- distributed_type: multi-GPU
|
| 46 |
+
- num_devices: 8
|
| 47 |
+
- gradient_accumulation_steps: 4
|
| 48 |
+
- total_train_batch_size: 128
|
| 49 |
+
- total_eval_batch_size: 32
|
| 50 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 51 |
+
- lr_scheduler_type: cosine
|
| 52 |
+
- lr_scheduler_warmup_ratio: 0.1
|
| 53 |
+
- num_epochs: 2
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 58 |
+
|:-------------:|:------:|:----:|:---------------:|
|
| 59 |
+
| 0.7442 | 0.9992 | 893 | 1.0789 |
|
| 60 |
+
| 0.6705 | 1.9983 | 1786 | 1.0630 |
|
| 61 |
+
|
| 62 |
+
|
| 63 |
+
### Framework versions
|
| 64 |
+
|
| 65 |
+
- Transformers 4.42.3
|
| 66 |
+
- Pytorch 2.1.2
|
| 67 |
+
- Datasets 2.20.0
|
| 68 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9983216783216782,
|
| 3 |
+
"total_flos": 1127451463778304.0,
|
| 4 |
+
"train_loss": 0.777597175032935,
|
| 5 |
+
"train_runtime": 16051.5455,
|
| 6 |
+
"train_samples": 1029132,
|
| 7 |
+
"train_samples_per_second": 14.253,
|
| 8 |
+
"train_steps_per_second": 0.111
|
| 9 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "HuggingFaceTB/SmolLM2-1.7B-8k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"LlamaForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "silu",
|
| 11 |
+
"hidden_size": 2048,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"intermediate_size": 8192,
|
| 14 |
+
"max_position_embeddings": 8192,
|
| 15 |
+
"mlp_bias": false,
|
| 16 |
+
"model_type": "llama",
|
| 17 |
+
"num_attention_heads": 32,
|
| 18 |
+
"num_hidden_layers": 24,
|
| 19 |
+
"num_key_value_heads": 32,
|
| 20 |
+
"pad_token_id": 2,
|
| 21 |
+
"pretraining_tp": 1,
|
| 22 |
+
"rms_norm_eps": 1e-05,
|
| 23 |
+
"rope_scaling": null,
|
| 24 |
+
"rope_theta": 130000,
|
| 25 |
+
"tie_word_embeddings": true,
|
| 26 |
+
"torch_dtype": "bfloat16",
|
| 27 |
+
"transformers_version": "4.42.3",
|
| 28 |
+
"use_cache": false,
|
| 29 |
+
"vocab_size": 49152
|
| 30 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 2,
|
| 6 |
+
"transformers_version": "4.42.3"
|
| 7 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:98360aebc58238dc07b4bcb78b709b7eb24d639e50a662f6bbdb57f610f6779c
|
| 3 |
+
size 3422777952
|
runs/Oct30_23-27-03_ip-26-0-164-18/events.out.tfevents.1730332615.ip-26-0-164-18.40006.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0873dd28e429f1d2206edf9d1303fb97e99dda399fea3d137c0e2cff6cda7703
|
| 3 |
+
size 81405
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"additional_special_tokens": [
|
| 3 |
+
{
|
| 4 |
+
"content": "<|im_start|>",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false
|
| 9 |
+
},
|
| 10 |
+
{
|
| 11 |
+
"content": "<|im_end|>",
|
| 12 |
+
"lstrip": false,
|
| 13 |
+
"normalized": false,
|
| 14 |
+
"rstrip": false,
|
| 15 |
+
"single_word": false
|
| 16 |
+
}
|
| 17 |
+
],
|
| 18 |
+
"bos_token": "<|im_start|>",
|
| 19 |
+
"eos_token": "<|im_end|>",
|
| 20 |
+
"pad_token": "<|im_end|>",
|
| 21 |
+
"unk_token": {
|
| 22 |
+
"content": "<|endoftext|>",
|
| 23 |
+
"lstrip": false,
|
| 24 |
+
"normalized": false,
|
| 25 |
+
"rstrip": false,
|
| 26 |
+
"single_word": false
|
| 27 |
+
}
|
| 28 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,154 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": false,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<|im_start|>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": false,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "<|im_end|>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": false,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<repo_name>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": false,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"4": {
|
| 37 |
+
"content": "<reponame>",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
},
|
| 44 |
+
"5": {
|
| 45 |
+
"content": "<file_sep>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"6": {
|
| 53 |
+
"content": "<filename>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"7": {
|
| 61 |
+
"content": "<gh_stars>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
},
|
| 68 |
+
"8": {
|
| 69 |
+
"content": "<issue_start>",
|
| 70 |
+
"lstrip": false,
|
| 71 |
+
"normalized": false,
|
| 72 |
+
"rstrip": false,
|
| 73 |
+
"single_word": false,
|
| 74 |
+
"special": true
|
| 75 |
+
},
|
| 76 |
+
"9": {
|
| 77 |
+
"content": "<issue_comment>",
|
| 78 |
+
"lstrip": false,
|
| 79 |
+
"normalized": false,
|
| 80 |
+
"rstrip": false,
|
| 81 |
+
"single_word": false,
|
| 82 |
+
"special": true
|
| 83 |
+
},
|
| 84 |
+
"10": {
|
| 85 |
+
"content": "<issue_closed>",
|
| 86 |
+
"lstrip": false,
|
| 87 |
+
"normalized": false,
|
| 88 |
+
"rstrip": false,
|
| 89 |
+
"single_word": false,
|
| 90 |
+
"special": true
|
| 91 |
+
},
|
| 92 |
+
"11": {
|
| 93 |
+
"content": "<jupyter_start>",
|
| 94 |
+
"lstrip": false,
|
| 95 |
+
"normalized": false,
|
| 96 |
+
"rstrip": false,
|
| 97 |
+
"single_word": false,
|
| 98 |
+
"special": true
|
| 99 |
+
},
|
| 100 |
+
"12": {
|
| 101 |
+
"content": "<jupyter_text>",
|
| 102 |
+
"lstrip": false,
|
| 103 |
+
"normalized": false,
|
| 104 |
+
"rstrip": false,
|
| 105 |
+
"single_word": false,
|
| 106 |
+
"special": true
|
| 107 |
+
},
|
| 108 |
+
"13": {
|
| 109 |
+
"content": "<jupyter_code>",
|
| 110 |
+
"lstrip": false,
|
| 111 |
+
"normalized": false,
|
| 112 |
+
"rstrip": false,
|
| 113 |
+
"single_word": false,
|
| 114 |
+
"special": true
|
| 115 |
+
},
|
| 116 |
+
"14": {
|
| 117 |
+
"content": "<jupyter_output>",
|
| 118 |
+
"lstrip": false,
|
| 119 |
+
"normalized": false,
|
| 120 |
+
"rstrip": false,
|
| 121 |
+
"single_word": false,
|
| 122 |
+
"special": true
|
| 123 |
+
},
|
| 124 |
+
"15": {
|
| 125 |
+
"content": "<jupyter_script>",
|
| 126 |
+
"lstrip": false,
|
| 127 |
+
"normalized": false,
|
| 128 |
+
"rstrip": false,
|
| 129 |
+
"single_word": false,
|
| 130 |
+
"special": true
|
| 131 |
+
},
|
| 132 |
+
"16": {
|
| 133 |
+
"content": "<empty_output>",
|
| 134 |
+
"lstrip": false,
|
| 135 |
+
"normalized": false,
|
| 136 |
+
"rstrip": false,
|
| 137 |
+
"single_word": false,
|
| 138 |
+
"special": true
|
| 139 |
+
}
|
| 140 |
+
},
|
| 141 |
+
"additional_special_tokens": [
|
| 142 |
+
"<|im_start|>",
|
| 143 |
+
"<|im_end|>"
|
| 144 |
+
],
|
| 145 |
+
"bos_token": "<|im_start|>",
|
| 146 |
+
"chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
|
| 147 |
+
"clean_up_tokenization_spaces": false,
|
| 148 |
+
"eos_token": "<|im_end|>",
|
| 149 |
+
"model_max_length": 2048,
|
| 150 |
+
"pad_token": "<|im_end|>",
|
| 151 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 152 |
+
"unk_token": "<|endoftext|>",
|
| 153 |
+
"vocab_size": 49152
|
| 154 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.9983216783216782,
|
| 3 |
+
"total_flos": 1127451463778304.0,
|
| 4 |
+
"train_loss": 0.777597175032935,
|
| 5 |
+
"train_runtime": 16051.5455,
|
| 6 |
+
"train_samples": 1029132,
|
| 7 |
+
"train_samples_per_second": 14.253,
|
| 8 |
+
"train_steps_per_second": 0.111
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2564 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9983216783216782,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 1786,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0011188811188811189,
|
| 13 |
+
"grad_norm": 153.8507844996399,
|
| 14 |
+
"learning_rate": 1.6759776536312848e-06,
|
| 15 |
+
"loss": 4.8778,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.005594405594405594,
|
| 20 |
+
"grad_norm": 153.39272858190785,
|
| 21 |
+
"learning_rate": 8.379888268156423e-06,
|
| 22 |
+
"loss": 4.7304,
|
| 23 |
+
"step": 5
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.011188811188811189,
|
| 27 |
+
"grad_norm": 96.41612092958107,
|
| 28 |
+
"learning_rate": 1.6759776536312845e-05,
|
| 29 |
+
"loss": 4.1396,
|
| 30 |
+
"step": 10
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.016783216783216783,
|
| 34 |
+
"grad_norm": 23.87392825650757,
|
| 35 |
+
"learning_rate": 2.513966480446927e-05,
|
| 36 |
+
"loss": 2.5431,
|
| 37 |
+
"step": 15
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.022377622377622378,
|
| 41 |
+
"grad_norm": 6.903016270261399,
|
| 42 |
+
"learning_rate": 3.351955307262569e-05,
|
| 43 |
+
"loss": 1.742,
|
| 44 |
+
"step": 20
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.027972027972027972,
|
| 48 |
+
"grad_norm": 2.441067831027445,
|
| 49 |
+
"learning_rate": 4.189944134078212e-05,
|
| 50 |
+
"loss": 1.3057,
|
| 51 |
+
"step": 25
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.033566433566433566,
|
| 55 |
+
"grad_norm": 1.0984763176398198,
|
| 56 |
+
"learning_rate": 5.027932960893854e-05,
|
| 57 |
+
"loss": 1.1221,
|
| 58 |
+
"step": 30
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.039160839160839164,
|
| 62 |
+
"grad_norm": 0.5734705027472714,
|
| 63 |
+
"learning_rate": 5.865921787709496e-05,
|
| 64 |
+
"loss": 1.0604,
|
| 65 |
+
"step": 35
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.044755244755244755,
|
| 69 |
+
"grad_norm": 0.47491071619847164,
|
| 70 |
+
"learning_rate": 6.703910614525138e-05,
|
| 71 |
+
"loss": 0.9951,
|
| 72 |
+
"step": 40
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.05034965034965035,
|
| 76 |
+
"grad_norm": 0.48157929747551576,
|
| 77 |
+
"learning_rate": 7.541899441340782e-05,
|
| 78 |
+
"loss": 0.9713,
|
| 79 |
+
"step": 45
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.055944055944055944,
|
| 83 |
+
"grad_norm": 0.45861937937875213,
|
| 84 |
+
"learning_rate": 8.379888268156423e-05,
|
| 85 |
+
"loss": 0.9551,
|
| 86 |
+
"step": 50
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.06153846153846154,
|
| 90 |
+
"grad_norm": 0.5047783392223855,
|
| 91 |
+
"learning_rate": 9.217877094972066e-05,
|
| 92 |
+
"loss": 0.948,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.06713286713286713,
|
| 97 |
+
"grad_norm": 0.4461540398939836,
|
| 98 |
+
"learning_rate": 0.00010055865921787709,
|
| 99 |
+
"loss": 0.9392,
|
| 100 |
+
"step": 60
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.07272727272727272,
|
| 104 |
+
"grad_norm": 0.3316401246824197,
|
| 105 |
+
"learning_rate": 0.00010893854748603351,
|
| 106 |
+
"loss": 0.9146,
|
| 107 |
+
"step": 65
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.07832167832167833,
|
| 111 |
+
"grad_norm": 0.21722066760341413,
|
| 112 |
+
"learning_rate": 0.00011731843575418992,
|
| 113 |
+
"loss": 0.9012,
|
| 114 |
+
"step": 70
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.08391608391608392,
|
| 118 |
+
"grad_norm": 0.1499443883049875,
|
| 119 |
+
"learning_rate": 0.00012569832402234635,
|
| 120 |
+
"loss": 0.8927,
|
| 121 |
+
"step": 75
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.08951048951048951,
|
| 125 |
+
"grad_norm": 0.22243490348769648,
|
| 126 |
+
"learning_rate": 0.00013407821229050276,
|
| 127 |
+
"loss": 0.8937,
|
| 128 |
+
"step": 80
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.0951048951048951,
|
| 132 |
+
"grad_norm": 0.37625345216705214,
|
| 133 |
+
"learning_rate": 0.0001424581005586592,
|
| 134 |
+
"loss": 0.8718,
|
| 135 |
+
"step": 85
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.1006993006993007,
|
| 139 |
+
"grad_norm": 0.24963786248117179,
|
| 140 |
+
"learning_rate": 0.00015083798882681564,
|
| 141 |
+
"loss": 0.8921,
|
| 142 |
+
"step": 90
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.1062937062937063,
|
| 146 |
+
"grad_norm": 0.26880921530140445,
|
| 147 |
+
"learning_rate": 0.00015921787709497208,
|
| 148 |
+
"loss": 0.8799,
|
| 149 |
+
"step": 95
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.11188811188811189,
|
| 153 |
+
"grad_norm": 0.20889087265244094,
|
| 154 |
+
"learning_rate": 0.00016759776536312847,
|
| 155 |
+
"loss": 0.8678,
|
| 156 |
+
"step": 100
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.11748251748251748,
|
| 160 |
+
"grad_norm": 0.20865989038699168,
|
| 161 |
+
"learning_rate": 0.0001759776536312849,
|
| 162 |
+
"loss": 0.8715,
|
| 163 |
+
"step": 105
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.12307692307692308,
|
| 167 |
+
"grad_norm": 0.17048448558431092,
|
| 168 |
+
"learning_rate": 0.00018435754189944132,
|
| 169 |
+
"loss": 0.866,
|
| 170 |
+
"step": 110
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.12867132867132866,
|
| 174 |
+
"grad_norm": 0.2563726823096772,
|
| 175 |
+
"learning_rate": 0.00019273743016759776,
|
| 176 |
+
"loss": 0.8786,
|
| 177 |
+
"step": 115
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.13426573426573427,
|
| 181 |
+
"grad_norm": 0.22854402328286066,
|
| 182 |
+
"learning_rate": 0.00020111731843575417,
|
| 183 |
+
"loss": 0.8641,
|
| 184 |
+
"step": 120
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.13986013986013987,
|
| 188 |
+
"grad_norm": 0.17876773160294948,
|
| 189 |
+
"learning_rate": 0.00020949720670391058,
|
| 190 |
+
"loss": 0.8625,
|
| 191 |
+
"step": 125
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.14545454545454545,
|
| 195 |
+
"grad_norm": 0.3615856863579962,
|
| 196 |
+
"learning_rate": 0.00021787709497206702,
|
| 197 |
+
"loss": 0.8504,
|
| 198 |
+
"step": 130
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.15104895104895105,
|
| 202 |
+
"grad_norm": 0.22751374781762473,
|
| 203 |
+
"learning_rate": 0.00022625698324022346,
|
| 204 |
+
"loss": 0.8683,
|
| 205 |
+
"step": 135
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.15664335664335666,
|
| 209 |
+
"grad_norm": 0.21790912834318224,
|
| 210 |
+
"learning_rate": 0.00023463687150837985,
|
| 211 |
+
"loss": 0.8519,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.16223776223776223,
|
| 216 |
+
"grad_norm": 0.18707627983249808,
|
| 217 |
+
"learning_rate": 0.0002430167597765363,
|
| 218 |
+
"loss": 0.8426,
|
| 219 |
+
"step": 145
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.16783216783216784,
|
| 223 |
+
"grad_norm": 0.22404831326656477,
|
| 224 |
+
"learning_rate": 0.0002513966480446927,
|
| 225 |
+
"loss": 0.8509,
|
| 226 |
+
"step": 150
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.17342657342657342,
|
| 230 |
+
"grad_norm": 0.20919591839479193,
|
| 231 |
+
"learning_rate": 0.00025977653631284914,
|
| 232 |
+
"loss": 0.8594,
|
| 233 |
+
"step": 155
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.17902097902097902,
|
| 237 |
+
"grad_norm": 0.15579954649232924,
|
| 238 |
+
"learning_rate": 0.0002681564245810055,
|
| 239 |
+
"loss": 0.8319,
|
| 240 |
+
"step": 160
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.18461538461538463,
|
| 244 |
+
"grad_norm": 0.4055853577053493,
|
| 245 |
+
"learning_rate": 0.00027653631284916196,
|
| 246 |
+
"loss": 0.8577,
|
| 247 |
+
"step": 165
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.1902097902097902,
|
| 251 |
+
"grad_norm": 0.2945253672148607,
|
| 252 |
+
"learning_rate": 0.0002849162011173184,
|
| 253 |
+
"loss": 0.8579,
|
| 254 |
+
"step": 170
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.1958041958041958,
|
| 258 |
+
"grad_norm": 0.21772142109085163,
|
| 259 |
+
"learning_rate": 0.00029329608938547484,
|
| 260 |
+
"loss": 0.8382,
|
| 261 |
+
"step": 175
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.2013986013986014,
|
| 265 |
+
"grad_norm": 0.2281215849497856,
|
| 266 |
+
"learning_rate": 0.00029999971336506766,
|
| 267 |
+
"loss": 0.8331,
|
| 268 |
+
"step": 180
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.206993006993007,
|
| 272 |
+
"grad_norm": 0.17899592128620762,
|
| 273 |
+
"learning_rate": 0.0002999896812574594,
|
| 274 |
+
"loss": 0.8434,
|
| 275 |
+
"step": 185
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.2125874125874126,
|
| 279 |
+
"grad_norm": 0.14731246899285452,
|
| 280 |
+
"learning_rate": 0.0002999653184986775,
|
| 281 |
+
"loss": 0.8458,
|
| 282 |
+
"step": 190
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.21818181818181817,
|
| 286 |
+
"grad_norm": 0.14945248170502354,
|
| 287 |
+
"learning_rate": 0.00029992662741644334,
|
| 288 |
+
"loss": 0.8457,
|
| 289 |
+
"step": 195
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.22377622377622378,
|
| 293 |
+
"grad_norm": 0.14124929357614965,
|
| 294 |
+
"learning_rate": 0.0002998736117074673,
|
| 295 |
+
"loss": 0.8219,
|
| 296 |
+
"step": 200
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.22937062937062938,
|
| 300 |
+
"grad_norm": 0.18621904528265473,
|
| 301 |
+
"learning_rate": 0.0002998062764370954,
|
| 302 |
+
"loss": 0.8299,
|
| 303 |
+
"step": 205
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.23496503496503496,
|
| 307 |
+
"grad_norm": 0.22513445720588005,
|
| 308 |
+
"learning_rate": 0.00029972462803882523,
|
| 309 |
+
"loss": 0.8502,
|
| 310 |
+
"step": 210
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.24055944055944056,
|
| 314 |
+
"grad_norm": 0.1636384983699745,
|
| 315 |
+
"learning_rate": 0.0002996286743136916,
|
| 316 |
+
"loss": 0.8506,
|
| 317 |
+
"step": 215
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.24615384615384617,
|
| 321 |
+
"grad_norm": 0.19587892370520038,
|
| 322 |
+
"learning_rate": 0.000299518424429521,
|
| 323 |
+
"loss": 0.8382,
|
| 324 |
+
"step": 220
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.2517482517482518,
|
| 328 |
+
"grad_norm": 0.17371559645370255,
|
| 329 |
+
"learning_rate": 0.0002993938889200556,
|
| 330 |
+
"loss": 0.8322,
|
| 331 |
+
"step": 225
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.2573426573426573,
|
| 335 |
+
"grad_norm": 0.13779589201692538,
|
| 336 |
+
"learning_rate": 0.0002992550796839468,
|
| 337 |
+
"loss": 0.8243,
|
| 338 |
+
"step": 230
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.2629370629370629,
|
| 342 |
+
"grad_norm": 0.22266694135489237,
|
| 343 |
+
"learning_rate": 0.00029910200998361857,
|
| 344 |
+
"loss": 0.8332,
|
| 345 |
+
"step": 235
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.26853146853146853,
|
| 349 |
+
"grad_norm": 0.21324201617886082,
|
| 350 |
+
"learning_rate": 0.0002989346944440003,
|
| 351 |
+
"loss": 0.8377,
|
| 352 |
+
"step": 240
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.27412587412587414,
|
| 356 |
+
"grad_norm": 0.1378342855087125,
|
| 357 |
+
"learning_rate": 0.0002987531490511291,
|
| 358 |
+
"loss": 0.8346,
|
| 359 |
+
"step": 245
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.27972027972027974,
|
| 363 |
+
"grad_norm": 0.13885067623091277,
|
| 364 |
+
"learning_rate": 0.000298557391150623,
|
| 365 |
+
"loss": 0.8366,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.2853146853146853,
|
| 370 |
+
"grad_norm": 0.13254675983004904,
|
| 371 |
+
"learning_rate": 0.00029834743944602316,
|
| 372 |
+
"loss": 0.8389,
|
| 373 |
+
"step": 255
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.2909090909090909,
|
| 377 |
+
"grad_norm": 0.15399189190157112,
|
| 378 |
+
"learning_rate": 0.0002981233139970071,
|
| 379 |
+
"loss": 0.8247,
|
| 380 |
+
"step": 260
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.2965034965034965,
|
| 384 |
+
"grad_norm": 0.16475565381996887,
|
| 385 |
+
"learning_rate": 0.0002978850362174722,
|
| 386 |
+
"loss": 0.8042,
|
| 387 |
+
"step": 265
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.3020979020979021,
|
| 391 |
+
"grad_norm": 0.1449584149552635,
|
| 392 |
+
"learning_rate": 0.0002976326288734894,
|
| 393 |
+
"loss": 0.8113,
|
| 394 |
+
"step": 270
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.3076923076923077,
|
| 398 |
+
"grad_norm": 0.1479148461740847,
|
| 399 |
+
"learning_rate": 0.0002973661160811284,
|
| 400 |
+
"loss": 0.8327,
|
| 401 |
+
"step": 275
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.3132867132867133,
|
| 405 |
+
"grad_norm": 0.11641111946130961,
|
| 406 |
+
"learning_rate": 0.00029708552330415337,
|
| 407 |
+
"loss": 0.8266,
|
| 408 |
+
"step": 280
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.31888111888111886,
|
| 412 |
+
"grad_norm": 0.1289644823631784,
|
| 413 |
+
"learning_rate": 0.0002967908773515898,
|
| 414 |
+
"loss": 0.8031,
|
| 415 |
+
"step": 285
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.32447552447552447,
|
| 419 |
+
"grad_norm": 0.19696780235530817,
|
| 420 |
+
"learning_rate": 0.0002964822063751635,
|
| 421 |
+
"loss": 0.8195,
|
| 422 |
+
"step": 290
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.3300699300699301,
|
| 426 |
+
"grad_norm": 0.13415245793361397,
|
| 427 |
+
"learning_rate": 0.00029615953986661056,
|
| 428 |
+
"loss": 0.8232,
|
| 429 |
+
"step": 295
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.3356643356643357,
|
| 433 |
+
"grad_norm": 0.1289064453230423,
|
| 434 |
+
"learning_rate": 0.0002958229086548595,
|
| 435 |
+
"loss": 0.811,
|
| 436 |
+
"step": 300
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.3412587412587413,
|
| 440 |
+
"grad_norm": 0.2030532789276685,
|
| 441 |
+
"learning_rate": 0.00029547234490308604,
|
| 442 |
+
"loss": 0.8196,
|
| 443 |
+
"step": 305
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.34685314685314683,
|
| 447 |
+
"grad_norm": 0.1690018504223368,
|
| 448 |
+
"learning_rate": 0.00029510788210563996,
|
| 449 |
+
"loss": 0.8176,
|
| 450 |
+
"step": 310
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.35244755244755244,
|
| 454 |
+
"grad_norm": 0.18348727298089326,
|
| 455 |
+
"learning_rate": 0.0002947295550848448,
|
| 456 |
+
"loss": 0.8106,
|
| 457 |
+
"step": 315
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.35804195804195804,
|
| 461 |
+
"grad_norm": 0.12176145136770436,
|
| 462 |
+
"learning_rate": 0.000294337399987671,
|
| 463 |
+
"loss": 0.8106,
|
| 464 |
+
"step": 320
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.36363636363636365,
|
| 468 |
+
"grad_norm": 0.14033633981287297,
|
| 469 |
+
"learning_rate": 0.0002939314542822821,
|
| 470 |
+
"loss": 0.8061,
|
| 471 |
+
"step": 325
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.36923076923076925,
|
| 475 |
+
"grad_norm": 0.1360601366236884,
|
| 476 |
+
"learning_rate": 0.0002935117567544547,
|
| 477 |
+
"loss": 0.8026,
|
| 478 |
+
"step": 330
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.3748251748251748,
|
| 482 |
+
"grad_norm": 0.15461610864168454,
|
| 483 |
+
"learning_rate": 0.0002930783475038734,
|
| 484 |
+
"loss": 0.8232,
|
| 485 |
+
"step": 335
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.3804195804195804,
|
| 489 |
+
"grad_norm": 0.10363695784053935,
|
| 490 |
+
"learning_rate": 0.0002926312679402985,
|
| 491 |
+
"loss": 0.8049,
|
| 492 |
+
"step": 340
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.386013986013986,
|
| 496 |
+
"grad_norm": 0.12894656658416323,
|
| 497 |
+
"learning_rate": 0.00029217056077961043,
|
| 498 |
+
"loss": 0.7993,
|
| 499 |
+
"step": 345
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.3916083916083916,
|
| 503 |
+
"grad_norm": 0.12013728582042035,
|
| 504 |
+
"learning_rate": 0.000291696270039728,
|
| 505 |
+
"loss": 0.7863,
|
| 506 |
+
"step": 350
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.3972027972027972,
|
| 510 |
+
"grad_norm": 0.10813698150493352,
|
| 511 |
+
"learning_rate": 0.0002912084410364029,
|
| 512 |
+
"loss": 0.7997,
|
| 513 |
+
"step": 355
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.4027972027972028,
|
| 517 |
+
"grad_norm": 0.17771298556035153,
|
| 518 |
+
"learning_rate": 0.00029070712037889,
|
| 519 |
+
"loss": 0.8018,
|
| 520 |
+
"step": 360
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.4083916083916084,
|
| 524 |
+
"grad_norm": 0.10837497819616222,
|
| 525 |
+
"learning_rate": 0.00029019235596549394,
|
| 526 |
+
"loss": 0.8078,
|
| 527 |
+
"step": 365
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.413986013986014,
|
| 531 |
+
"grad_norm": 0.12560903325827824,
|
| 532 |
+
"learning_rate": 0.0002896641969789932,
|
| 533 |
+
"loss": 0.8182,
|
| 534 |
+
"step": 370
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.4195804195804196,
|
| 538 |
+
"grad_norm": 0.17753966157649756,
|
| 539 |
+
"learning_rate": 0.0002891226938819405,
|
| 540 |
+
"loss": 0.8059,
|
| 541 |
+
"step": 375
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.4251748251748252,
|
| 545 |
+
"grad_norm": 0.12975220832490364,
|
| 546 |
+
"learning_rate": 0.0002885678984118415,
|
| 547 |
+
"loss": 0.7811,
|
| 548 |
+
"step": 380
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.4307692307692308,
|
| 552 |
+
"grad_norm": 0.18552799892105348,
|
| 553 |
+
"learning_rate": 0.0002879998635762118,
|
| 554 |
+
"loss": 0.799,
|
| 555 |
+
"step": 385
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.43636363636363634,
|
| 559 |
+
"grad_norm": 0.17744047449094297,
|
| 560 |
+
"learning_rate": 0.000287418643647512,
|
| 561 |
+
"loss": 0.7974,
|
| 562 |
+
"step": 390
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.44195804195804195,
|
| 566 |
+
"grad_norm": 0.13797503272598644,
|
| 567 |
+
"learning_rate": 0.00028682429415796267,
|
| 568 |
+
"loss": 0.7931,
|
| 569 |
+
"step": 395
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.44755244755244755,
|
| 573 |
+
"grad_norm": 0.12860800142514783,
|
| 574 |
+
"learning_rate": 0.0002862168718942383,
|
| 575 |
+
"loss": 0.7861,
|
| 576 |
+
"step": 400
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.45314685314685316,
|
| 580 |
+
"grad_norm": 0.12668805301774957,
|
| 581 |
+
"learning_rate": 0.00028559643489204186,
|
| 582 |
+
"loss": 0.8107,
|
| 583 |
+
"step": 405
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.45874125874125876,
|
| 587 |
+
"grad_norm": 0.11338400624283074,
|
| 588 |
+
"learning_rate": 0.0002849630424305595,
|
| 589 |
+
"loss": 0.8088,
|
| 590 |
+
"step": 410
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.4643356643356643,
|
| 594 |
+
"grad_norm": 0.1345570809457518,
|
| 595 |
+
"learning_rate": 0.00028431675502679717,
|
| 596 |
+
"loss": 0.8038,
|
| 597 |
+
"step": 415
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.4699300699300699,
|
| 601 |
+
"grad_norm": 0.11102734283999488,
|
| 602 |
+
"learning_rate": 0.00028365763442979823,
|
| 603 |
+
"loss": 0.8163,
|
| 604 |
+
"step": 420
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.4755244755244755,
|
| 608 |
+
"grad_norm": 0.12810877098380058,
|
| 609 |
+
"learning_rate": 0.000282985743614744,
|
| 610 |
+
"loss": 0.8017,
|
| 611 |
+
"step": 425
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.4811188811188811,
|
| 615 |
+
"grad_norm": 0.11873107897366113,
|
| 616 |
+
"learning_rate": 0.0002823011467769364,
|
| 617 |
+
"loss": 0.7957,
|
| 618 |
+
"step": 430
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.48671328671328673,
|
| 622 |
+
"grad_norm": 0.13272846461187762,
|
| 623 |
+
"learning_rate": 0.000281603909325665,
|
| 624 |
+
"loss": 0.8054,
|
| 625 |
+
"step": 435
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.49230769230769234,
|
| 629 |
+
"grad_norm": 0.10875049751356179,
|
| 630 |
+
"learning_rate": 0.00028089409787795716,
|
| 631 |
+
"loss": 0.7976,
|
| 632 |
+
"step": 440
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.4979020979020979,
|
| 636 |
+
"grad_norm": 0.11634952268956057,
|
| 637 |
+
"learning_rate": 0.0002801717802522132,
|
| 638 |
+
"loss": 0.792,
|
| 639 |
+
"step": 445
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 0.5034965034965035,
|
| 643 |
+
"grad_norm": 0.10607280522202725,
|
| 644 |
+
"learning_rate": 0.00027943702546172697,
|
| 645 |
+
"loss": 0.8078,
|
| 646 |
+
"step": 450
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 0.509090909090909,
|
| 650 |
+
"grad_norm": 0.15901805386261234,
|
| 651 |
+
"learning_rate": 0.00027868990370809164,
|
| 652 |
+
"loss": 0.8023,
|
| 653 |
+
"step": 455
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 0.5146853146853146,
|
| 657 |
+
"grad_norm": 0.12222508621029372,
|
| 658 |
+
"learning_rate": 0.00027793048637449273,
|
| 659 |
+
"loss": 0.7956,
|
| 660 |
+
"step": 460
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 0.5202797202797202,
|
| 664 |
+
"grad_norm": 0.10769602993732762,
|
| 665 |
+
"learning_rate": 0.0002771588460188876,
|
| 666 |
+
"loss": 0.7897,
|
| 667 |
+
"step": 465
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 0.5258741258741259,
|
| 671 |
+
"grad_norm": 0.1119724845850892,
|
| 672 |
+
"learning_rate": 0.00027637505636707315,
|
| 673 |
+
"loss": 0.7901,
|
| 674 |
+
"step": 470
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.5314685314685315,
|
| 678 |
+
"grad_norm": 0.11279764446201575,
|
| 679 |
+
"learning_rate": 0.0002755791923056415,
|
| 680 |
+
"loss": 0.79,
|
| 681 |
+
"step": 475
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 0.5370629370629371,
|
| 685 |
+
"grad_norm": 0.16791268697388115,
|
| 686 |
+
"learning_rate": 0.0002747713298748253,
|
| 687 |
+
"loss": 0.7909,
|
| 688 |
+
"step": 480
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 0.5426573426573427,
|
| 692 |
+
"grad_norm": 0.14084691053374204,
|
| 693 |
+
"learning_rate": 0.00027395154626123225,
|
| 694 |
+
"loss": 0.8013,
|
| 695 |
+
"step": 485
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 0.5482517482517483,
|
| 699 |
+
"grad_norm": 0.12452861008817544,
|
| 700 |
+
"learning_rate": 0.00027311991979047046,
|
| 701 |
+
"loss": 0.7888,
|
| 702 |
+
"step": 490
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 0.5538461538461539,
|
| 706 |
+
"grad_norm": 0.12432408060887955,
|
| 707 |
+
"learning_rate": 0.00027227652991966507,
|
| 708 |
+
"loss": 0.7736,
|
| 709 |
+
"step": 495
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 0.5594405594405595,
|
| 713 |
+
"grad_norm": 0.1167273022435753,
|
| 714 |
+
"learning_rate": 0.00027142145722986637,
|
| 715 |
+
"loss": 0.7892,
|
| 716 |
+
"step": 500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.5650349650349651,
|
| 720 |
+
"grad_norm": 0.1570515101772784,
|
| 721 |
+
"learning_rate": 0.0002705547834183506,
|
| 722 |
+
"loss": 0.7735,
|
| 723 |
+
"step": 505
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 0.5706293706293706,
|
| 727 |
+
"grad_norm": 0.1757157276572827,
|
| 728 |
+
"learning_rate": 0.00026967659129081465,
|
| 729 |
+
"loss": 0.7947,
|
| 730 |
+
"step": 510
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 0.5762237762237762,
|
| 734 |
+
"grad_norm": 0.10842613803532425,
|
| 735 |
+
"learning_rate": 0.0002687869647534643,
|
| 736 |
+
"loss": 0.7844,
|
| 737 |
+
"step": 515
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 0.5818181818181818,
|
| 741 |
+
"grad_norm": 0.14559223171878447,
|
| 742 |
+
"learning_rate": 0.0002678859888049972,
|
| 743 |
+
"loss": 0.7881,
|
| 744 |
+
"step": 520
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 0.5874125874125874,
|
| 748 |
+
"grad_norm": 0.13398356861671987,
|
| 749 |
+
"learning_rate": 0.0002669737495284819,
|
| 750 |
+
"loss": 0.7731,
|
| 751 |
+
"step": 525
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 0.593006993006993,
|
| 755 |
+
"grad_norm": 0.16426998549603838,
|
| 756 |
+
"learning_rate": 0.00026605033408313354,
|
| 757 |
+
"loss": 0.7819,
|
| 758 |
+
"step": 530
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.5986013986013986,
|
| 762 |
+
"grad_norm": 0.12765279377973468,
|
| 763 |
+
"learning_rate": 0.0002651158306959855,
|
| 764 |
+
"loss": 0.7725,
|
| 765 |
+
"step": 535
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 0.6041958041958042,
|
| 769 |
+
"grad_norm": 0.11028117529632633,
|
| 770 |
+
"learning_rate": 0.00026417032865346023,
|
| 771 |
+
"loss": 0.7926,
|
| 772 |
+
"step": 540
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 0.6097902097902098,
|
| 776 |
+
"grad_norm": 0.12782289803306612,
|
| 777 |
+
"learning_rate": 0.00026321391829283884,
|
| 778 |
+
"loss": 0.7634,
|
| 779 |
+
"step": 545
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 0.6153846153846154,
|
| 783 |
+
"grad_norm": 0.1369549911517866,
|
| 784 |
+
"learning_rate": 0.0002622466909936289,
|
| 785 |
+
"loss": 0.7628,
|
| 786 |
+
"step": 550
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 0.620979020979021,
|
| 790 |
+
"grad_norm": 0.11521467825966626,
|
| 791 |
+
"learning_rate": 0.0002612687391688347,
|
| 792 |
+
"loss": 0.7763,
|
| 793 |
+
"step": 555
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 0.6265734265734266,
|
| 797 |
+
"grad_norm": 0.12350512688772212,
|
| 798 |
+
"learning_rate": 0.00026028015625612706,
|
| 799 |
+
"loss": 0.7884,
|
| 800 |
+
"step": 560
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.6321678321678321,
|
| 804 |
+
"grad_norm": 0.10825882258061882,
|
| 805 |
+
"learning_rate": 0.000259281036708916,
|
| 806 |
+
"loss": 0.7945,
|
| 807 |
+
"step": 565
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 0.6377622377622377,
|
| 811 |
+
"grad_norm": 0.11150400278758185,
|
| 812 |
+
"learning_rate": 0.00025827147598732656,
|
| 813 |
+
"loss": 0.7862,
|
| 814 |
+
"step": 570
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 0.6433566433566433,
|
| 818 |
+
"grad_norm": 0.10337975508652891,
|
| 819 |
+
"learning_rate": 0.00025725157054907777,
|
| 820 |
+
"loss": 0.7838,
|
| 821 |
+
"step": 575
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 0.6489510489510489,
|
| 825 |
+
"grad_norm": 0.09770680702107988,
|
| 826 |
+
"learning_rate": 0.0002562214178402669,
|
| 827 |
+
"loss": 0.7969,
|
| 828 |
+
"step": 580
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 0.6545454545454545,
|
| 832 |
+
"grad_norm": 0.1163323874465344,
|
| 833 |
+
"learning_rate": 0.00025518111628605885,
|
| 834 |
+
"loss": 0.7819,
|
| 835 |
+
"step": 585
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 0.6601398601398601,
|
| 839 |
+
"grad_norm": 0.11896295851189105,
|
| 840 |
+
"learning_rate": 0.00025413076528128255,
|
| 841 |
+
"loss": 0.7709,
|
| 842 |
+
"step": 590
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.6657342657342658,
|
| 846 |
+
"grad_norm": 0.11330549108690711,
|
| 847 |
+
"learning_rate": 0.0002530704651809339,
|
| 848 |
+
"loss": 0.7744,
|
| 849 |
+
"step": 595
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 0.6713286713286714,
|
| 853 |
+
"grad_norm": 0.13654029100152085,
|
| 854 |
+
"learning_rate": 0.0002520003172905878,
|
| 855 |
+
"loss": 0.7952,
|
| 856 |
+
"step": 600
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 0.676923076923077,
|
| 860 |
+
"grad_norm": 0.10351068732360345,
|
| 861 |
+
"learning_rate": 0.0002509204238567186,
|
| 862 |
+
"loss": 0.7755,
|
| 863 |
+
"step": 605
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 0.6825174825174826,
|
| 867 |
+
"grad_norm": 0.14077884569885882,
|
| 868 |
+
"learning_rate": 0.00024983088805693163,
|
| 869 |
+
"loss": 0.7831,
|
| 870 |
+
"step": 610
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 0.6881118881118881,
|
| 874 |
+
"grad_norm": 0.11948510265909233,
|
| 875 |
+
"learning_rate": 0.00024873181399010446,
|
| 876 |
+
"loss": 0.7861,
|
| 877 |
+
"step": 615
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 0.6937062937062937,
|
| 881 |
+
"grad_norm": 0.12755481254736167,
|
| 882 |
+
"learning_rate": 0.00024762330666644136,
|
| 883 |
+
"loss": 0.7782,
|
| 884 |
+
"step": 620
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.6993006993006993,
|
| 888 |
+
"grad_norm": 0.10102773025347063,
|
| 889 |
+
"learning_rate": 0.0002465054719974401,
|
| 890 |
+
"loss": 0.7731,
|
| 891 |
+
"step": 625
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 0.7048951048951049,
|
| 895 |
+
"grad_norm": 0.11665875780674151,
|
| 896 |
+
"learning_rate": 0.0002453784167857725,
|
| 897 |
+
"loss": 0.7839,
|
| 898 |
+
"step": 630
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 0.7104895104895105,
|
| 902 |
+
"grad_norm": 0.1074509342573267,
|
| 903 |
+
"learning_rate": 0.00024424224871508014,
|
| 904 |
+
"loss": 0.7769,
|
| 905 |
+
"step": 635
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 0.7160839160839161,
|
| 909 |
+
"grad_norm": 0.12422918258396683,
|
| 910 |
+
"learning_rate": 0.0002430970763396861,
|
| 911 |
+
"loss": 0.7754,
|
| 912 |
+
"step": 640
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 0.7216783216783217,
|
| 916 |
+
"grad_norm": 0.11786940518440715,
|
| 917 |
+
"learning_rate": 0.00024194300907422276,
|
| 918 |
+
"loss": 0.7974,
|
| 919 |
+
"step": 645
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 0.7272727272727273,
|
| 923 |
+
"grad_norm": 0.10398253596299759,
|
| 924 |
+
"learning_rate": 0.00024078015718317818,
|
| 925 |
+
"loss": 0.7729,
|
| 926 |
+
"step": 650
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.7328671328671329,
|
| 930 |
+
"grad_norm": 0.09903776559603776,
|
| 931 |
+
"learning_rate": 0.00023960863177036079,
|
| 932 |
+
"loss": 0.774,
|
| 933 |
+
"step": 655
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 0.7384615384615385,
|
| 937 |
+
"grad_norm": 0.09179018818178636,
|
| 938 |
+
"learning_rate": 0.00023842854476828411,
|
| 939 |
+
"loss": 0.7629,
|
| 940 |
+
"step": 660
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 0.7440559440559441,
|
| 944 |
+
"grad_norm": 0.10143244330106851,
|
| 945 |
+
"learning_rate": 0.0002372400089274724,
|
| 946 |
+
"loss": 0.781,
|
| 947 |
+
"step": 665
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 0.7496503496503496,
|
| 951 |
+
"grad_norm": 0.1121174501554115,
|
| 952 |
+
"learning_rate": 0.00023604313780568772,
|
| 953 |
+
"loss": 0.7811,
|
| 954 |
+
"step": 670
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 0.7552447552447552,
|
| 958 |
+
"grad_norm": 0.10047620860251714,
|
| 959 |
+
"learning_rate": 0.00023483804575708027,
|
| 960 |
+
"loss": 0.7752,
|
| 961 |
+
"step": 675
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 0.7608391608391608,
|
| 965 |
+
"grad_norm": 0.11068987100435401,
|
| 966 |
+
"learning_rate": 0.0002336248479212626,
|
| 967 |
+
"loss": 0.7657,
|
| 968 |
+
"step": 680
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.7664335664335664,
|
| 972 |
+
"grad_norm": 0.12535370469312143,
|
| 973 |
+
"learning_rate": 0.0002324036602123086,
|
| 974 |
+
"loss": 0.7731,
|
| 975 |
+
"step": 685
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 0.772027972027972,
|
| 979 |
+
"grad_norm": 0.11059830294321518,
|
| 980 |
+
"learning_rate": 0.00023117459930767847,
|
| 981 |
+
"loss": 0.7831,
|
| 982 |
+
"step": 690
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 0.7776223776223776,
|
| 986 |
+
"grad_norm": 0.11073279493368117,
|
| 987 |
+
"learning_rate": 0.00022993778263707105,
|
| 988 |
+
"loss": 0.7705,
|
| 989 |
+
"step": 695
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 0.7832167832167832,
|
| 993 |
+
"grad_norm": 0.12764331660853,
|
| 994 |
+
"learning_rate": 0.000228693328371204,
|
| 995 |
+
"loss": 0.7816,
|
| 996 |
+
"step": 700
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 0.7888111888111888,
|
| 1000 |
+
"grad_norm": 0.10951855988735575,
|
| 1001 |
+
"learning_rate": 0.0002274413554105232,
|
| 1002 |
+
"loss": 0.7577,
|
| 1003 |
+
"step": 705
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 0.7944055944055944,
|
| 1007 |
+
"grad_norm": 0.12320185793067193,
|
| 1008 |
+
"learning_rate": 0.00022618198337384264,
|
| 1009 |
+
"loss": 0.7744,
|
| 1010 |
+
"step": 710
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.8,
|
| 1014 |
+
"grad_norm": 0.10716655164464266,
|
| 1015 |
+
"learning_rate": 0.00022491533258691546,
|
| 1016 |
+
"loss": 0.7752,
|
| 1017 |
+
"step": 715
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 0.8055944055944056,
|
| 1021 |
+
"grad_norm": 0.11962877035421185,
|
| 1022 |
+
"learning_rate": 0.00022364152407093737,
|
| 1023 |
+
"loss": 0.7812,
|
| 1024 |
+
"step": 720
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 0.8111888111888111,
|
| 1028 |
+
"grad_norm": 0.10651252517324981,
|
| 1029 |
+
"learning_rate": 0.00022236067953098414,
|
| 1030 |
+
"loss": 0.78,
|
| 1031 |
+
"step": 725
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 0.8167832167832167,
|
| 1035 |
+
"grad_norm": 0.1086890422520719,
|
| 1036 |
+
"learning_rate": 0.00022107292134438298,
|
| 1037 |
+
"loss": 0.7801,
|
| 1038 |
+
"step": 730
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 0.8223776223776224,
|
| 1042 |
+
"grad_norm": 0.09743032182612474,
|
| 1043 |
+
"learning_rate": 0.00021977837254902034,
|
| 1044 |
+
"loss": 0.7762,
|
| 1045 |
+
"step": 735
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 0.827972027972028,
|
| 1049 |
+
"grad_norm": 0.09964450561189496,
|
| 1050 |
+
"learning_rate": 0.0002184771568315862,
|
| 1051 |
+
"loss": 0.7809,
|
| 1052 |
+
"step": 740
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 0.8335664335664336,
|
| 1056 |
+
"grad_norm": 0.09852105354101268,
|
| 1057 |
+
"learning_rate": 0.0002171693985157567,
|
| 1058 |
+
"loss": 0.7803,
|
| 1059 |
+
"step": 745
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 0.8391608391608392,
|
| 1063 |
+
"grad_norm": 0.10751247181443387,
|
| 1064 |
+
"learning_rate": 0.00021585522255031554,
|
| 1065 |
+
"loss": 0.754,
|
| 1066 |
+
"step": 750
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 0.8447552447552448,
|
| 1070 |
+
"grad_norm": 0.10083658202350868,
|
| 1071 |
+
"learning_rate": 0.00021453475449721593,
|
| 1072 |
+
"loss": 0.7689,
|
| 1073 |
+
"step": 755
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 0.8503496503496504,
|
| 1077 |
+
"grad_norm": 0.1132302634053956,
|
| 1078 |
+
"learning_rate": 0.00021320812051958392,
|
| 1079 |
+
"loss": 0.7667,
|
| 1080 |
+
"step": 760
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 0.855944055944056,
|
| 1084 |
+
"grad_norm": 0.09942757452904256,
|
| 1085 |
+
"learning_rate": 0.00021187544736966403,
|
| 1086 |
+
"loss": 0.7798,
|
| 1087 |
+
"step": 765
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 0.8615384615384616,
|
| 1091 |
+
"grad_norm": 0.11119180151464317,
|
| 1092 |
+
"learning_rate": 0.00021053686237670912,
|
| 1093 |
+
"loss": 0.7768,
|
| 1094 |
+
"step": 770
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 0.8671328671328671,
|
| 1098 |
+
"grad_norm": 0.11012292213675227,
|
| 1099 |
+
"learning_rate": 0.0002091924934348146,
|
| 1100 |
+
"loss": 0.7641,
|
| 1101 |
+
"step": 775
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 0.8727272727272727,
|
| 1105 |
+
"grad_norm": 0.10508899250762296,
|
| 1106 |
+
"learning_rate": 0.0002078424689906988,
|
| 1107 |
+
"loss": 0.772,
|
| 1108 |
+
"step": 780
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 0.8783216783216783,
|
| 1112 |
+
"grad_norm": 0.09916724236606267,
|
| 1113 |
+
"learning_rate": 0.00020648691803143088,
|
| 1114 |
+
"loss": 0.7798,
|
| 1115 |
+
"step": 785
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 0.8839160839160839,
|
| 1119 |
+
"grad_norm": 0.10139136991498161,
|
| 1120 |
+
"learning_rate": 0.00020512597007210672,
|
| 1121 |
+
"loss": 0.7595,
|
| 1122 |
+
"step": 790
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 0.8895104895104895,
|
| 1126 |
+
"grad_norm": 0.09922677378640356,
|
| 1127 |
+
"learning_rate": 0.00020375975514347447,
|
| 1128 |
+
"loss": 0.7582,
|
| 1129 |
+
"step": 795
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 0.8951048951048951,
|
| 1133 |
+
"grad_norm": 0.10734407771584646,
|
| 1134 |
+
"learning_rate": 0.0002023884037795109,
|
| 1135 |
+
"loss": 0.7747,
|
| 1136 |
+
"step": 800
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 0.9006993006993007,
|
| 1140 |
+
"grad_norm": 0.09003146154158007,
|
| 1141 |
+
"learning_rate": 0.00020101204700494963,
|
| 1142 |
+
"loss": 0.772,
|
| 1143 |
+
"step": 805
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 0.9062937062937063,
|
| 1147 |
+
"grad_norm": 0.10224227095976963,
|
| 1148 |
+
"learning_rate": 0.00019963081632276244,
|
| 1149 |
+
"loss": 0.7632,
|
| 1150 |
+
"step": 810
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 0.9118881118881119,
|
| 1154 |
+
"grad_norm": 0.10634964894349899,
|
| 1155 |
+
"learning_rate": 0.00019824484370159511,
|
| 1156 |
+
"loss": 0.7621,
|
| 1157 |
+
"step": 815
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 0.9174825174825175,
|
| 1161 |
+
"grad_norm": 0.08669203171955091,
|
| 1162 |
+
"learning_rate": 0.00019685426156315817,
|
| 1163 |
+
"loss": 0.7678,
|
| 1164 |
+
"step": 820
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 0.9230769230769231,
|
| 1168 |
+
"grad_norm": 0.09236536065178236,
|
| 1169 |
+
"learning_rate": 0.00019545920276957512,
|
| 1170 |
+
"loss": 0.7615,
|
| 1171 |
+
"step": 825
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 0.9286713286713286,
|
| 1175 |
+
"grad_norm": 0.09051106183048227,
|
| 1176 |
+
"learning_rate": 0.00019405980061068813,
|
| 1177 |
+
"loss": 0.7538,
|
| 1178 |
+
"step": 830
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 0.9342657342657342,
|
| 1182 |
+
"grad_norm": 0.09075314123364801,
|
| 1183 |
+
"learning_rate": 0.00019265618879132294,
|
| 1184 |
+
"loss": 0.7695,
|
| 1185 |
+
"step": 835
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 0.9398601398601398,
|
| 1189 |
+
"grad_norm": 0.09779655862085175,
|
| 1190 |
+
"learning_rate": 0.000191248501418514,
|
| 1191 |
+
"loss": 0.7486,
|
| 1192 |
+
"step": 840
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 0.9454545454545454,
|
| 1196 |
+
"grad_norm": 0.09284439899834651,
|
| 1197 |
+
"learning_rate": 0.00018983687298869165,
|
| 1198 |
+
"loss": 0.7757,
|
| 1199 |
+
"step": 845
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 0.951048951048951,
|
| 1203 |
+
"grad_norm": 0.10906263614923466,
|
| 1204 |
+
"learning_rate": 0.00018842143837483137,
|
| 1205 |
+
"loss": 0.7654,
|
| 1206 |
+
"step": 850
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 0.9566433566433566,
|
| 1210 |
+
"grad_norm": 0.08711563033296012,
|
| 1211 |
+
"learning_rate": 0.00018700233281356774,
|
| 1212 |
+
"loss": 0.7661,
|
| 1213 |
+
"step": 855
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 0.9622377622377623,
|
| 1217 |
+
"grad_norm": 0.10105386217514606,
|
| 1218 |
+
"learning_rate": 0.00018557969189227327,
|
| 1219 |
+
"loss": 0.7566,
|
| 1220 |
+
"step": 860
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 0.9678321678321679,
|
| 1224 |
+
"grad_norm": 0.09374075010702233,
|
| 1225 |
+
"learning_rate": 0.00018415365153610363,
|
| 1226 |
+
"loss": 0.7505,
|
| 1227 |
+
"step": 865
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 0.9734265734265735,
|
| 1231 |
+
"grad_norm": 0.09217739631223294,
|
| 1232 |
+
"learning_rate": 0.00018272434799501108,
|
| 1233 |
+
"loss": 0.7513,
|
| 1234 |
+
"step": 870
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 0.9790209790209791,
|
| 1238 |
+
"grad_norm": 0.10860300198578388,
|
| 1239 |
+
"learning_rate": 0.00018129191783072644,
|
| 1240 |
+
"loss": 0.7586,
|
| 1241 |
+
"step": 875
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 0.9846153846153847,
|
| 1245 |
+
"grad_norm": 0.10382198085691989,
|
| 1246 |
+
"learning_rate": 0.00017985649790371123,
|
| 1247 |
+
"loss": 0.7712,
|
| 1248 |
+
"step": 880
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 0.9902097902097902,
|
| 1252 |
+
"grad_norm": 0.09718012392096963,
|
| 1253 |
+
"learning_rate": 0.00017841822536008174,
|
| 1254 |
+
"loss": 0.7548,
|
| 1255 |
+
"step": 885
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 0.9958041958041958,
|
| 1259 |
+
"grad_norm": 0.09129380340752612,
|
| 1260 |
+
"learning_rate": 0.00017697723761850529,
|
| 1261 |
+
"loss": 0.7442,
|
| 1262 |
+
"step": 890
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 0.9991608391608392,
|
| 1266 |
+
"eval_loss": 1.078917145729065,
|
| 1267 |
+
"eval_runtime": 368.4649,
|
| 1268 |
+
"eval_samples_per_second": 55.878,
|
| 1269 |
+
"eval_steps_per_second": 1.748,
|
| 1270 |
+
"step": 893
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 1.0013986013986014,
|
| 1274 |
+
"grad_norm": 0.2193429577704184,
|
| 1275 |
+
"learning_rate": 0.0001755336723570709,
|
| 1276 |
+
"loss": 0.7304,
|
| 1277 |
+
"step": 895
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 1.006993006993007,
|
| 1281 |
+
"grad_norm": 0.12029696024418647,
|
| 1282 |
+
"learning_rate": 0.00017408766750013455,
|
| 1283 |
+
"loss": 0.6883,
|
| 1284 |
+
"step": 900
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 1.0125874125874126,
|
| 1288 |
+
"grad_norm": 0.11157245101821324,
|
| 1289 |
+
"learning_rate": 0.0001726393612051416,
|
| 1290 |
+
"loss": 0.6937,
|
| 1291 |
+
"step": 905
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 1.018181818181818,
|
| 1295 |
+
"grad_norm": 0.11562847307206518,
|
| 1296 |
+
"learning_rate": 0.0001711888918494268,
|
| 1297 |
+
"loss": 0.7072,
|
| 1298 |
+
"step": 910
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 1.0237762237762238,
|
| 1302 |
+
"grad_norm": 0.09699091401508494,
|
| 1303 |
+
"learning_rate": 0.00016973639801699258,
|
| 1304 |
+
"loss": 0.7002,
|
| 1305 |
+
"step": 915
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 1.0293706293706293,
|
| 1309 |
+
"grad_norm": 0.10329864513789806,
|
| 1310 |
+
"learning_rate": 0.0001682820184852687,
|
| 1311 |
+
"loss": 0.7049,
|
| 1312 |
+
"step": 920
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 1.034965034965035,
|
| 1316 |
+
"grad_norm": 0.10505439144371424,
|
| 1317 |
+
"learning_rate": 0.0001668258922118525,
|
| 1318 |
+
"loss": 0.7062,
|
| 1319 |
+
"step": 925
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 1.0405594405594405,
|
| 1323 |
+
"grad_norm": 0.09345853743066072,
|
| 1324 |
+
"learning_rate": 0.0001653681583212326,
|
| 1325 |
+
"loss": 0.705,
|
| 1326 |
+
"step": 930
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 1.0461538461538462,
|
| 1330 |
+
"grad_norm": 0.14624259074546206,
|
| 1331 |
+
"learning_rate": 0.00016390895609149608,
|
| 1332 |
+
"loss": 0.6862,
|
| 1333 |
+
"step": 935
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 1.0517482517482517,
|
| 1337 |
+
"grad_norm": 0.09484870970062163,
|
| 1338 |
+
"learning_rate": 0.00016244842494102135,
|
| 1339 |
+
"loss": 0.6794,
|
| 1340 |
+
"step": 940
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 1.0573426573426574,
|
| 1344 |
+
"grad_norm": 0.1093126433741654,
|
| 1345 |
+
"learning_rate": 0.00016098670441515759,
|
| 1346 |
+
"loss": 0.6965,
|
| 1347 |
+
"step": 945
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 1.062937062937063,
|
| 1351 |
+
"grad_norm": 0.08709955416922696,
|
| 1352 |
+
"learning_rate": 0.000159523934172892,
|
| 1353 |
+
"loss": 0.6875,
|
| 1354 |
+
"step": 950
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 1.0685314685314686,
|
| 1358 |
+
"grad_norm": 0.08587078936550793,
|
| 1359 |
+
"learning_rate": 0.00015806025397350617,
|
| 1360 |
+
"loss": 0.6816,
|
| 1361 |
+
"step": 955
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 1.0741258741258741,
|
| 1365 |
+
"grad_norm": 0.09226255228367097,
|
| 1366 |
+
"learning_rate": 0.00015659580366322265,
|
| 1367 |
+
"loss": 0.6909,
|
| 1368 |
+
"step": 960
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 1.0797202797202796,
|
| 1372 |
+
"grad_norm": 0.0936240844549708,
|
| 1373 |
+
"learning_rate": 0.00015513072316184393,
|
| 1374 |
+
"loss": 0.6904,
|
| 1375 |
+
"step": 965
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 1.0853146853146853,
|
| 1379 |
+
"grad_norm": 0.085659439411888,
|
| 1380 |
+
"learning_rate": 0.0001536651524493834,
|
| 1381 |
+
"loss": 0.6874,
|
| 1382 |
+
"step": 970
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 1.0909090909090908,
|
| 1386 |
+
"grad_norm": 0.09924612907304702,
|
| 1387 |
+
"learning_rate": 0.00015219923155269157,
|
| 1388 |
+
"loss": 0.6953,
|
| 1389 |
+
"step": 975
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 1.0965034965034965,
|
| 1393 |
+
"grad_norm": 0.10028513455623837,
|
| 1394 |
+
"learning_rate": 0.00015073310053207665,
|
| 1395 |
+
"loss": 0.6967,
|
| 1396 |
+
"step": 980
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 1.102097902097902,
|
| 1400 |
+
"grad_norm": 0.09345136457275896,
|
| 1401 |
+
"learning_rate": 0.00014926689946792332,
|
| 1402 |
+
"loss": 0.6905,
|
| 1403 |
+
"step": 985
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 1.1076923076923078,
|
| 1407 |
+
"grad_norm": 0.08893419418399162,
|
| 1408 |
+
"learning_rate": 0.00014780076844730849,
|
| 1409 |
+
"loss": 0.6985,
|
| 1410 |
+
"step": 990
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 1.1132867132867132,
|
| 1414 |
+
"grad_norm": 0.08853702003030435,
|
| 1415 |
+
"learning_rate": 0.00014633484755061658,
|
| 1416 |
+
"loss": 0.7014,
|
| 1417 |
+
"step": 995
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 1.118881118881119,
|
| 1421 |
+
"grad_norm": 0.08365464476596743,
|
| 1422 |
+
"learning_rate": 0.0001448692768381561,
|
| 1423 |
+
"loss": 0.697,
|
| 1424 |
+
"step": 1000
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 1.1244755244755245,
|
| 1428 |
+
"grad_norm": 0.08780213786162074,
|
| 1429 |
+
"learning_rate": 0.00014340419633677732,
|
| 1430 |
+
"loss": 0.7025,
|
| 1431 |
+
"step": 1005
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 1.1300699300699302,
|
| 1435 |
+
"grad_norm": 0.0884295613746686,
|
| 1436 |
+
"learning_rate": 0.00014193974602649386,
|
| 1437 |
+
"loss": 0.6993,
|
| 1438 |
+
"step": 1010
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 1.1356643356643357,
|
| 1442 |
+
"grad_norm": 0.09059540370700289,
|
| 1443 |
+
"learning_rate": 0.00014047606582710798,
|
| 1444 |
+
"loss": 0.6948,
|
| 1445 |
+
"step": 1015
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 1.1412587412587412,
|
| 1449 |
+
"grad_norm": 0.09899100050319447,
|
| 1450 |
+
"learning_rate": 0.00013901329558484236,
|
| 1451 |
+
"loss": 0.6992,
|
| 1452 |
+
"step": 1020
|
| 1453 |
+
},
|
| 1454 |
+
{
|
| 1455 |
+
"epoch": 1.1468531468531469,
|
| 1456 |
+
"grad_norm": 0.09283620847093584,
|
| 1457 |
+
"learning_rate": 0.00013755157505897868,
|
| 1458 |
+
"loss": 0.7184,
|
| 1459 |
+
"step": 1025
|
| 1460 |
+
},
|
| 1461 |
+
{
|
| 1462 |
+
"epoch": 1.1524475524475524,
|
| 1463 |
+
"grad_norm": 0.08887024389956383,
|
| 1464 |
+
"learning_rate": 0.00013609104390850392,
|
| 1465 |
+
"loss": 0.697,
|
| 1466 |
+
"step": 1030
|
| 1467 |
+
},
|
| 1468 |
+
{
|
| 1469 |
+
"epoch": 1.158041958041958,
|
| 1470 |
+
"grad_norm": 0.08693816785751791,
|
| 1471 |
+
"learning_rate": 0.0001346318416787674,
|
| 1472 |
+
"loss": 0.6939,
|
| 1473 |
+
"step": 1035
|
| 1474 |
+
},
|
| 1475 |
+
{
|
| 1476 |
+
"epoch": 1.1636363636363636,
|
| 1477 |
+
"grad_norm": 0.09116425941497776,
|
| 1478 |
+
"learning_rate": 0.00013317410778814745,
|
| 1479 |
+
"loss": 0.6989,
|
| 1480 |
+
"step": 1040
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 1.1692307692307693,
|
| 1484 |
+
"grad_norm": 0.08570466903745366,
|
| 1485 |
+
"learning_rate": 0.00013171798151473133,
|
| 1486 |
+
"loss": 0.6956,
|
| 1487 |
+
"step": 1045
|
| 1488 |
+
},
|
| 1489 |
+
{
|
| 1490 |
+
"epoch": 1.1748251748251748,
|
| 1491 |
+
"grad_norm": 0.0974948891545614,
|
| 1492 |
+
"learning_rate": 0.0001302636019830074,
|
| 1493 |
+
"loss": 0.6965,
|
| 1494 |
+
"step": 1050
|
| 1495 |
+
},
|
| 1496 |
+
{
|
| 1497 |
+
"epoch": 1.1804195804195805,
|
| 1498 |
+
"grad_norm": 0.08995597994573838,
|
| 1499 |
+
"learning_rate": 0.0001288111081505732,
|
| 1500 |
+
"loss": 0.7041,
|
| 1501 |
+
"step": 1055
|
| 1502 |
+
},
|
| 1503 |
+
{
|
| 1504 |
+
"epoch": 1.186013986013986,
|
| 1505 |
+
"grad_norm": 0.07861264078439004,
|
| 1506 |
+
"learning_rate": 0.00012736063879485837,
|
| 1507 |
+
"loss": 0.7032,
|
| 1508 |
+
"step": 1060
|
| 1509 |
+
},
|
| 1510 |
+
{
|
| 1511 |
+
"epoch": 1.1916083916083915,
|
| 1512 |
+
"grad_norm": 0.0841514205254347,
|
| 1513 |
+
"learning_rate": 0.0001259123324998655,
|
| 1514 |
+
"loss": 0.6905,
|
| 1515 |
+
"step": 1065
|
| 1516 |
+
},
|
| 1517 |
+
{
|
| 1518 |
+
"epoch": 1.1972027972027972,
|
| 1519 |
+
"grad_norm": 0.08692540683602883,
|
| 1520 |
+
"learning_rate": 0.0001244663276429291,
|
| 1521 |
+
"loss": 0.7074,
|
| 1522 |
+
"step": 1070
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 1.2027972027972027,
|
| 1526 |
+
"grad_norm": 0.08776833500890557,
|
| 1527 |
+
"learning_rate": 0.00012302276238149463,
|
| 1528 |
+
"loss": 0.7041,
|
| 1529 |
+
"step": 1075
|
| 1530 |
+
},
|
| 1531 |
+
{
|
| 1532 |
+
"epoch": 1.2083916083916084,
|
| 1533 |
+
"grad_norm": 0.08727180482054969,
|
| 1534 |
+
"learning_rate": 0.00012158177463991828,
|
| 1535 |
+
"loss": 0.696,
|
| 1536 |
+
"step": 1080
|
| 1537 |
+
},
|
| 1538 |
+
{
|
| 1539 |
+
"epoch": 1.213986013986014,
|
| 1540 |
+
"grad_norm": 0.08046559487267867,
|
| 1541 |
+
"learning_rate": 0.00012014350209628875,
|
| 1542 |
+
"loss": 0.6826,
|
| 1543 |
+
"step": 1085
|
| 1544 |
+
},
|
| 1545 |
+
{
|
| 1546 |
+
"epoch": 1.2195804195804196,
|
| 1547 |
+
"grad_norm": 0.08781752670322365,
|
| 1548 |
+
"learning_rate": 0.00011870808216927356,
|
| 1549 |
+
"loss": 0.6999,
|
| 1550 |
+
"step": 1090
|
| 1551 |
+
},
|
| 1552 |
+
{
|
| 1553 |
+
"epoch": 1.2251748251748251,
|
| 1554 |
+
"grad_norm": 0.08813740501797863,
|
| 1555 |
+
"learning_rate": 0.00011727565200498888,
|
| 1556 |
+
"loss": 0.7037,
|
| 1557 |
+
"step": 1095
|
| 1558 |
+
},
|
| 1559 |
+
{
|
| 1560 |
+
"epoch": 1.2307692307692308,
|
| 1561 |
+
"grad_norm": 0.08772222997454243,
|
| 1562 |
+
"learning_rate": 0.00011584634846389638,
|
| 1563 |
+
"loss": 0.6986,
|
| 1564 |
+
"step": 1100
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 1.2363636363636363,
|
| 1568 |
+
"grad_norm": 0.10332929019104932,
|
| 1569 |
+
"learning_rate": 0.00011442030810772673,
|
| 1570 |
+
"loss": 0.6725,
|
| 1571 |
+
"step": 1105
|
| 1572 |
+
},
|
| 1573 |
+
{
|
| 1574 |
+
"epoch": 1.241958041958042,
|
| 1575 |
+
"grad_norm": 0.08709301562019137,
|
| 1576 |
+
"learning_rate": 0.00011299766718643226,
|
| 1577 |
+
"loss": 0.7063,
|
| 1578 |
+
"step": 1110
|
| 1579 |
+
},
|
| 1580 |
+
{
|
| 1581 |
+
"epoch": 1.2475524475524475,
|
| 1582 |
+
"grad_norm": 0.08449696490718896,
|
| 1583 |
+
"learning_rate": 0.00011157856162516863,
|
| 1584 |
+
"loss": 0.692,
|
| 1585 |
+
"step": 1115
|
| 1586 |
+
},
|
| 1587 |
+
{
|
| 1588 |
+
"epoch": 1.2531468531468533,
|
| 1589 |
+
"grad_norm": 0.08124616270666368,
|
| 1590 |
+
"learning_rate": 0.00011016312701130841,
|
| 1591 |
+
"loss": 0.6915,
|
| 1592 |
+
"step": 1120
|
| 1593 |
+
},
|
| 1594 |
+
{
|
| 1595 |
+
"epoch": 1.2587412587412588,
|
| 1596 |
+
"grad_norm": 0.09376360999458151,
|
| 1597 |
+
"learning_rate": 0.000108751498581486,
|
| 1598 |
+
"loss": 0.6939,
|
| 1599 |
+
"step": 1125
|
| 1600 |
+
},
|
| 1601 |
+
{
|
| 1602 |
+
"epoch": 1.2643356643356642,
|
| 1603 |
+
"grad_norm": 0.08677694965635109,
|
| 1604 |
+
"learning_rate": 0.00010734381120867707,
|
| 1605 |
+
"loss": 0.7029,
|
| 1606 |
+
"step": 1130
|
| 1607 |
+
},
|
| 1608 |
+
{
|
| 1609 |
+
"epoch": 1.26993006993007,
|
| 1610 |
+
"grad_norm": 0.08737879433308637,
|
| 1611 |
+
"learning_rate": 0.00010594019938931187,
|
| 1612 |
+
"loss": 0.6849,
|
| 1613 |
+
"step": 1135
|
| 1614 |
+
},
|
| 1615 |
+
{
|
| 1616 |
+
"epoch": 1.2755244755244755,
|
| 1617 |
+
"grad_norm": 0.08311271942623545,
|
| 1618 |
+
"learning_rate": 0.00010454079723042485,
|
| 1619 |
+
"loss": 0.6799,
|
| 1620 |
+
"step": 1140
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 1.2811188811188812,
|
| 1624 |
+
"grad_norm": 0.07508666997892824,
|
| 1625 |
+
"learning_rate": 0.00010314573843684183,
|
| 1626 |
+
"loss": 0.6979,
|
| 1627 |
+
"step": 1145
|
| 1628 |
+
},
|
| 1629 |
+
{
|
| 1630 |
+
"epoch": 1.2867132867132867,
|
| 1631 |
+
"grad_norm": 0.07967922610661686,
|
| 1632 |
+
"learning_rate": 0.00010175515629840487,
|
| 1633 |
+
"loss": 0.6793,
|
| 1634 |
+
"step": 1150
|
| 1635 |
+
},
|
| 1636 |
+
{
|
| 1637 |
+
"epoch": 1.2923076923076924,
|
| 1638 |
+
"grad_norm": 0.08373250104545919,
|
| 1639 |
+
"learning_rate": 0.00010036918367723754,
|
| 1640 |
+
"loss": 0.6942,
|
| 1641 |
+
"step": 1155
|
| 1642 |
+
},
|
| 1643 |
+
{
|
| 1644 |
+
"epoch": 1.2979020979020979,
|
| 1645 |
+
"grad_norm": 0.08415435083106365,
|
| 1646 |
+
"learning_rate": 9.898795299505037e-05,
|
| 1647 |
+
"loss": 0.6843,
|
| 1648 |
+
"step": 1160
|
| 1649 |
+
},
|
| 1650 |
+
{
|
| 1651 |
+
"epoch": 1.3034965034965036,
|
| 1652 |
+
"grad_norm": 0.0972925237131085,
|
| 1653 |
+
"learning_rate": 9.761159622048914e-05,
|
| 1654 |
+
"loss": 0.6786,
|
| 1655 |
+
"step": 1165
|
| 1656 |
+
},
|
| 1657 |
+
{
|
| 1658 |
+
"epoch": 1.309090909090909,
|
| 1659 |
+
"grad_norm": 0.08332589008281578,
|
| 1660 |
+
"learning_rate": 9.624024485652552e-05,
|
| 1661 |
+
"loss": 0.6895,
|
| 1662 |
+
"step": 1170
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 1.3146853146853146,
|
| 1666 |
+
"grad_norm": 0.08324352673680815,
|
| 1667 |
+
"learning_rate": 9.48740299278933e-05,
|
| 1668 |
+
"loss": 0.6902,
|
| 1669 |
+
"step": 1175
|
| 1670 |
+
},
|
| 1671 |
+
{
|
| 1672 |
+
"epoch": 1.3202797202797203,
|
| 1673 |
+
"grad_norm": 0.08212510567034546,
|
| 1674 |
+
"learning_rate": 9.351308196856911e-05,
|
| 1675 |
+
"loss": 0.6861,
|
| 1676 |
+
"step": 1180
|
| 1677 |
+
},
|
| 1678 |
+
{
|
| 1679 |
+
"epoch": 1.325874125874126,
|
| 1680 |
+
"grad_norm": 0.08320001036155726,
|
| 1681 |
+
"learning_rate": 9.215753100930118e-05,
|
| 1682 |
+
"loss": 0.6943,
|
| 1683 |
+
"step": 1185
|
| 1684 |
+
},
|
| 1685 |
+
{
|
| 1686 |
+
"epoch": 1.3314685314685315,
|
| 1687 |
+
"grad_norm": 0.09026366102833903,
|
| 1688 |
+
"learning_rate": 9.08075065651854e-05,
|
| 1689 |
+
"loss": 0.7031,
|
| 1690 |
+
"step": 1190
|
| 1691 |
+
},
|
| 1692 |
+
{
|
| 1693 |
+
"epoch": 1.337062937062937,
|
| 1694 |
+
"grad_norm": 0.07809171125830346,
|
| 1695 |
+
"learning_rate": 8.946313762329081e-05,
|
| 1696 |
+
"loss": 0.6974,
|
| 1697 |
+
"step": 1195
|
| 1698 |
+
},
|
| 1699 |
+
{
|
| 1700 |
+
"epoch": 1.3426573426573427,
|
| 1701 |
+
"grad_norm": 0.07708050011669573,
|
| 1702 |
+
"learning_rate": 8.812455263033595e-05,
|
| 1703 |
+
"loss": 0.7072,
|
| 1704 |
+
"step": 1200
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 1.3482517482517482,
|
| 1708 |
+
"grad_norm": 0.08672106026682487,
|
| 1709 |
+
"learning_rate": 8.679187948041605e-05,
|
| 1710 |
+
"loss": 0.6946,
|
| 1711 |
+
"step": 1205
|
| 1712 |
+
},
|
| 1713 |
+
{
|
| 1714 |
+
"epoch": 1.353846153846154,
|
| 1715 |
+
"grad_norm": 0.08223294476361295,
|
| 1716 |
+
"learning_rate": 8.546524550278405e-05,
|
| 1717 |
+
"loss": 0.6917,
|
| 1718 |
+
"step": 1210
|
| 1719 |
+
},
|
| 1720 |
+
{
|
| 1721 |
+
"epoch": 1.3594405594405594,
|
| 1722 |
+
"grad_norm": 0.07572041983819904,
|
| 1723 |
+
"learning_rate": 8.414477744968441e-05,
|
| 1724 |
+
"loss": 0.7068,
|
| 1725 |
+
"step": 1215
|
| 1726 |
+
},
|
| 1727 |
+
{
|
| 1728 |
+
"epoch": 1.365034965034965,
|
| 1729 |
+
"grad_norm": 0.07935310328973877,
|
| 1730 |
+
"learning_rate": 8.283060148424328e-05,
|
| 1731 |
+
"loss": 0.6825,
|
| 1732 |
+
"step": 1220
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 1.3706293706293706,
|
| 1736 |
+
"grad_norm": 0.08084222220036809,
|
| 1737 |
+
"learning_rate": 8.152284316841382e-05,
|
| 1738 |
+
"loss": 0.6895,
|
| 1739 |
+
"step": 1225
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 1.3762237762237763,
|
| 1743 |
+
"grad_norm": 0.08362754142551204,
|
| 1744 |
+
"learning_rate": 8.02216274509797e-05,
|
| 1745 |
+
"loss": 0.6855,
|
| 1746 |
+
"step": 1230
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 1.3818181818181818,
|
| 1750 |
+
"grad_norm": 0.08263146843817087,
|
| 1751 |
+
"learning_rate": 7.892707865561702e-05,
|
| 1752 |
+
"loss": 0.685,
|
| 1753 |
+
"step": 1235
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 1.3874125874125873,
|
| 1757 |
+
"grad_norm": 0.07614085795637057,
|
| 1758 |
+
"learning_rate": 7.763932046901587e-05,
|
| 1759 |
+
"loss": 0.698,
|
| 1760 |
+
"step": 1240
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 1.393006993006993,
|
| 1764 |
+
"grad_norm": 0.08297289392137173,
|
| 1765 |
+
"learning_rate": 7.635847592906259e-05,
|
| 1766 |
+
"loss": 0.6892,
|
| 1767 |
+
"step": 1245
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 1.3986013986013985,
|
| 1771 |
+
"grad_norm": 0.09257478488971461,
|
| 1772 |
+
"learning_rate": 7.50846674130845e-05,
|
| 1773 |
+
"loss": 0.6819,
|
| 1774 |
+
"step": 1250
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 1.4041958041958043,
|
| 1778 |
+
"grad_norm": 0.0820635181024322,
|
| 1779 |
+
"learning_rate": 7.381801662615731e-05,
|
| 1780 |
+
"loss": 0.6836,
|
| 1781 |
+
"step": 1255
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 1.4097902097902097,
|
| 1785 |
+
"grad_norm": 0.08479286795375468,
|
| 1786 |
+
"learning_rate": 7.255864458947677e-05,
|
| 1787 |
+
"loss": 0.6838,
|
| 1788 |
+
"step": 1260
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 1.4153846153846155,
|
| 1792 |
+
"grad_norm": 0.08169299364006637,
|
| 1793 |
+
"learning_rate": 7.130667162879602e-05,
|
| 1794 |
+
"loss": 0.6912,
|
| 1795 |
+
"step": 1265
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 1.420979020979021,
|
| 1799 |
+
"grad_norm": 0.0847296311273153,
|
| 1800 |
+
"learning_rate": 7.006221736292892e-05,
|
| 1801 |
+
"loss": 0.6824,
|
| 1802 |
+
"step": 1270
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 1.4265734265734267,
|
| 1806 |
+
"grad_norm": 0.07999623959693618,
|
| 1807 |
+
"learning_rate": 6.882540069232155e-05,
|
| 1808 |
+
"loss": 0.6806,
|
| 1809 |
+
"step": 1275
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 1.4321678321678322,
|
| 1813 |
+
"grad_norm": 0.07964123970317083,
|
| 1814 |
+
"learning_rate": 6.759633978769139e-05,
|
| 1815 |
+
"loss": 0.7052,
|
| 1816 |
+
"step": 1280
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 1.4377622377622377,
|
| 1820 |
+
"grad_norm": 0.0966871184419301,
|
| 1821 |
+
"learning_rate": 6.63751520787374e-05,
|
| 1822 |
+
"loss": 0.7002,
|
| 1823 |
+
"step": 1285
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 1.4433566433566434,
|
| 1827 |
+
"grad_norm": 0.09626897693462537,
|
| 1828 |
+
"learning_rate": 6.516195424291972e-05,
|
| 1829 |
+
"loss": 0.6912,
|
| 1830 |
+
"step": 1290
|
| 1831 |
+
},
|
| 1832 |
+
{
|
| 1833 |
+
"epoch": 1.4489510489510489,
|
| 1834 |
+
"grad_norm": 0.08145120964067751,
|
| 1835 |
+
"learning_rate": 6.395686219431232e-05,
|
| 1836 |
+
"loss": 0.6877,
|
| 1837 |
+
"step": 1295
|
| 1838 |
+
},
|
| 1839 |
+
{
|
| 1840 |
+
"epoch": 1.4545454545454546,
|
| 1841 |
+
"grad_norm": 0.08789566288442652,
|
| 1842 |
+
"learning_rate": 6.275999107252758e-05,
|
| 1843 |
+
"loss": 0.6847,
|
| 1844 |
+
"step": 1300
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 1.46013986013986,
|
| 1848 |
+
"grad_norm": 0.0806679485875012,
|
| 1849 |
+
"learning_rate": 6.157145523171587e-05,
|
| 1850 |
+
"loss": 0.6869,
|
| 1851 |
+
"step": 1305
|
| 1852 |
+
},
|
| 1853 |
+
{
|
| 1854 |
+
"epoch": 1.4657342657342658,
|
| 1855 |
+
"grad_norm": 0.07942983212018809,
|
| 1856 |
+
"learning_rate": 6.039136822963924e-05,
|
| 1857 |
+
"loss": 0.6767,
|
| 1858 |
+
"step": 1310
|
| 1859 |
+
},
|
| 1860 |
+
{
|
| 1861 |
+
"epoch": 1.4713286713286713,
|
| 1862 |
+
"grad_norm": 0.08489122954397245,
|
| 1863 |
+
"learning_rate": 5.9219842816821796e-05,
|
| 1864 |
+
"loss": 0.6814,
|
| 1865 |
+
"step": 1315
|
| 1866 |
+
},
|
| 1867 |
+
{
|
| 1868 |
+
"epoch": 1.476923076923077,
|
| 1869 |
+
"grad_norm": 0.08653144402466852,
|
| 1870 |
+
"learning_rate": 5.805699092577722e-05,
|
| 1871 |
+
"loss": 0.6968,
|
| 1872 |
+
"step": 1320
|
| 1873 |
+
},
|
| 1874 |
+
{
|
| 1875 |
+
"epoch": 1.4825174825174825,
|
| 1876 |
+
"grad_norm": 0.08218462407837061,
|
| 1877 |
+
"learning_rate": 5.6902923660313855e-05,
|
| 1878 |
+
"loss": 0.6781,
|
| 1879 |
+
"step": 1325
|
| 1880 |
+
},
|
| 1881 |
+
{
|
| 1882 |
+
"epoch": 1.488111888111888,
|
| 1883 |
+
"grad_norm": 0.07605468398853794,
|
| 1884 |
+
"learning_rate": 5.5757751284919836e-05,
|
| 1885 |
+
"loss": 0.6837,
|
| 1886 |
+
"step": 1330
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 1.4937062937062937,
|
| 1890 |
+
"grad_norm": 0.07347540817224318,
|
| 1891 |
+
"learning_rate": 5.462158321422751e-05,
|
| 1892 |
+
"loss": 0.678,
|
| 1893 |
+
"step": 1335
|
| 1894 |
+
},
|
| 1895 |
+
{
|
| 1896 |
+
"epoch": 1.4993006993006994,
|
| 1897 |
+
"grad_norm": 0.08136761352155132,
|
| 1898 |
+
"learning_rate": 5.34945280025599e-05,
|
| 1899 |
+
"loss": 0.6843,
|
| 1900 |
+
"step": 1340
|
| 1901 |
+
},
|
| 1902 |
+
{
|
| 1903 |
+
"epoch": 1.504895104895105,
|
| 1904 |
+
"grad_norm": 0.079793665685405,
|
| 1905 |
+
"learning_rate": 5.237669333355863e-05,
|
| 1906 |
+
"loss": 0.6919,
|
| 1907 |
+
"step": 1345
|
| 1908 |
+
},
|
| 1909 |
+
{
|
| 1910 |
+
"epoch": 1.5104895104895104,
|
| 1911 |
+
"grad_norm": 0.07721773411337415,
|
| 1912 |
+
"learning_rate": 5.126818600989557e-05,
|
| 1913 |
+
"loss": 0.6826,
|
| 1914 |
+
"step": 1350
|
| 1915 |
+
},
|
| 1916 |
+
{
|
| 1917 |
+
"epoch": 1.5160839160839161,
|
| 1918 |
+
"grad_norm": 0.07622519698946323,
|
| 1919 |
+
"learning_rate": 5.0169111943068374e-05,
|
| 1920 |
+
"loss": 0.6901,
|
| 1921 |
+
"step": 1355
|
| 1922 |
+
},
|
| 1923 |
+
{
|
| 1924 |
+
"epoch": 1.5216783216783218,
|
| 1925 |
+
"grad_norm": 0.07862095262493349,
|
| 1926 |
+
"learning_rate": 4.9079576143281326e-05,
|
| 1927 |
+
"loss": 0.691,
|
| 1928 |
+
"step": 1360
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 1.5272727272727273,
|
| 1932 |
+
"grad_norm": 0.07949120411421966,
|
| 1933 |
+
"learning_rate": 4.7999682709412216e-05,
|
| 1934 |
+
"loss": 0.6806,
|
| 1935 |
+
"step": 1365
|
| 1936 |
+
},
|
| 1937 |
+
{
|
| 1938 |
+
"epoch": 1.5328671328671328,
|
| 1939 |
+
"grad_norm": 0.08530944602336236,
|
| 1940 |
+
"learning_rate": 4.692953481906605e-05,
|
| 1941 |
+
"loss": 0.6847,
|
| 1942 |
+
"step": 1370
|
| 1943 |
+
},
|
| 1944 |
+
{
|
| 1945 |
+
"epoch": 1.5384615384615383,
|
| 1946 |
+
"grad_norm": 0.08074798934800313,
|
| 1947 |
+
"learning_rate": 4.586923471871743e-05,
|
| 1948 |
+
"loss": 0.681,
|
| 1949 |
+
"step": 1375
|
| 1950 |
+
},
|
| 1951 |
+
{
|
| 1952 |
+
"epoch": 1.544055944055944,
|
| 1953 |
+
"grad_norm": 0.0777449518216711,
|
| 1954 |
+
"learning_rate": 4.481888371394115e-05,
|
| 1955 |
+
"loss": 0.6874,
|
| 1956 |
+
"step": 1380
|
| 1957 |
+
},
|
| 1958 |
+
{
|
| 1959 |
+
"epoch": 1.5496503496503498,
|
| 1960 |
+
"grad_norm": 0.07535968734459726,
|
| 1961 |
+
"learning_rate": 4.377858215973318e-05,
|
| 1962 |
+
"loss": 0.6751,
|
| 1963 |
+
"step": 1385
|
| 1964 |
+
},
|
| 1965 |
+
{
|
| 1966 |
+
"epoch": 1.5552447552447553,
|
| 1967 |
+
"grad_norm": 0.07188075306681382,
|
| 1968 |
+
"learning_rate": 4.2748429450922263e-05,
|
| 1969 |
+
"loss": 0.6745,
|
| 1970 |
+
"step": 1390
|
| 1971 |
+
},
|
| 1972 |
+
{
|
| 1973 |
+
"epoch": 1.5608391608391607,
|
| 1974 |
+
"grad_norm": 0.0781312708212868,
|
| 1975 |
+
"learning_rate": 4.172852401267347e-05,
|
| 1976 |
+
"loss": 0.688,
|
| 1977 |
+
"step": 1395
|
| 1978 |
+
},
|
| 1979 |
+
{
|
| 1980 |
+
"epoch": 1.5664335664335665,
|
| 1981 |
+
"grad_norm": 0.07521827314744757,
|
| 1982 |
+
"learning_rate": 4.0718963291084e-05,
|
| 1983 |
+
"loss": 0.6757,
|
| 1984 |
+
"step": 1400
|
| 1985 |
+
},
|
| 1986 |
+
{
|
| 1987 |
+
"epoch": 1.5720279720279722,
|
| 1988 |
+
"grad_norm": 0.07373545234599518,
|
| 1989 |
+
"learning_rate": 3.9719843743872964e-05,
|
| 1990 |
+
"loss": 0.6778,
|
| 1991 |
+
"step": 1405
|
| 1992 |
+
},
|
| 1993 |
+
{
|
| 1994 |
+
"epoch": 1.5776223776223777,
|
| 1995 |
+
"grad_norm": 0.07264270828709123,
|
| 1996 |
+
"learning_rate": 3.873126083116525e-05,
|
| 1997 |
+
"loss": 0.6864,
|
| 1998 |
+
"step": 1410
|
| 1999 |
+
},
|
| 2000 |
+
{
|
| 2001 |
+
"epoch": 1.5832167832167832,
|
| 2002 |
+
"grad_norm": 0.0763271584430483,
|
| 2003 |
+
"learning_rate": 3.775330900637108e-05,
|
| 2004 |
+
"loss": 0.683,
|
| 2005 |
+
"step": 1415
|
| 2006 |
+
},
|
| 2007 |
+
{
|
| 2008 |
+
"epoch": 1.5888111888111887,
|
| 2009 |
+
"grad_norm": 0.07981572030031307,
|
| 2010 |
+
"learning_rate": 3.678608170716117e-05,
|
| 2011 |
+
"loss": 0.6795,
|
| 2012 |
+
"step": 1420
|
| 2013 |
+
},
|
| 2014 |
+
{
|
| 2015 |
+
"epoch": 1.5944055944055944,
|
| 2016 |
+
"grad_norm": 0.0746896810608302,
|
| 2017 |
+
"learning_rate": 3.582967134653972e-05,
|
| 2018 |
+
"loss": 0.675,
|
| 2019 |
+
"step": 1425
|
| 2020 |
+
},
|
| 2021 |
+
{
|
| 2022 |
+
"epoch": 1.6,
|
| 2023 |
+
"grad_norm": 0.07449703423407898,
|
| 2024 |
+
"learning_rate": 3.488416930401457e-05,
|
| 2025 |
+
"loss": 0.6805,
|
| 2026 |
+
"step": 1430
|
| 2027 |
+
},
|
| 2028 |
+
{
|
| 2029 |
+
"epoch": 1.6055944055944056,
|
| 2030 |
+
"grad_norm": 0.07104343697673421,
|
| 2031 |
+
"learning_rate": 3.3949665916866466e-05,
|
| 2032 |
+
"loss": 0.6752,
|
| 2033 |
+
"step": 1435
|
| 2034 |
+
},
|
| 2035 |
+
{
|
| 2036 |
+
"epoch": 1.611188811188811,
|
| 2037 |
+
"grad_norm": 0.07498548014087274,
|
| 2038 |
+
"learning_rate": 3.302625047151807e-05,
|
| 2039 |
+
"loss": 0.6949,
|
| 2040 |
+
"step": 1440
|
| 2041 |
+
},
|
| 2042 |
+
{
|
| 2043 |
+
"epoch": 1.6167832167832168,
|
| 2044 |
+
"grad_norm": 0.07539631481012679,
|
| 2045 |
+
"learning_rate": 3.211401119500283e-05,
|
| 2046 |
+
"loss": 0.6892,
|
| 2047 |
+
"step": 1445
|
| 2048 |
+
},
|
| 2049 |
+
{
|
| 2050 |
+
"epoch": 1.6223776223776225,
|
| 2051 |
+
"grad_norm": 0.0765200306903301,
|
| 2052 |
+
"learning_rate": 3.12130352465357e-05,
|
| 2053 |
+
"loss": 0.6891,
|
| 2054 |
+
"step": 1450
|
| 2055 |
+
},
|
| 2056 |
+
{
|
| 2057 |
+
"epoch": 1.627972027972028,
|
| 2058 |
+
"grad_norm": 0.0730441214772004,
|
| 2059 |
+
"learning_rate": 3.032340870918527e-05,
|
| 2060 |
+
"loss": 0.6981,
|
| 2061 |
+
"step": 1455
|
| 2062 |
+
},
|
| 2063 |
+
{
|
| 2064 |
+
"epoch": 1.6335664335664335,
|
| 2065 |
+
"grad_norm": 0.08171074628055888,
|
| 2066 |
+
"learning_rate": 2.9445216581649384e-05,
|
| 2067 |
+
"loss": 0.6936,
|
| 2068 |
+
"step": 1460
|
| 2069 |
+
},
|
| 2070 |
+
{
|
| 2071 |
+
"epoch": 1.6391608391608392,
|
| 2072 |
+
"grad_norm": 0.083112396316528,
|
| 2073 |
+
"learning_rate": 2.8578542770133654e-05,
|
| 2074 |
+
"loss": 0.6737,
|
| 2075 |
+
"step": 1465
|
| 2076 |
+
},
|
| 2077 |
+
{
|
| 2078 |
+
"epoch": 1.6447552447552447,
|
| 2079 |
+
"grad_norm": 0.08252295941296424,
|
| 2080 |
+
"learning_rate": 2.772347008033492e-05,
|
| 2081 |
+
"loss": 0.701,
|
| 2082 |
+
"step": 1470
|
| 2083 |
+
},
|
| 2084 |
+
{
|
| 2085 |
+
"epoch": 1.6503496503496504,
|
| 2086 |
+
"grad_norm": 0.07835293344801121,
|
| 2087 |
+
"learning_rate": 2.688008020952952e-05,
|
| 2088 |
+
"loss": 0.6921,
|
| 2089 |
+
"step": 1475
|
| 2090 |
+
},
|
| 2091 |
+
{
|
| 2092 |
+
"epoch": 1.655944055944056,
|
| 2093 |
+
"grad_norm": 0.0778212904803543,
|
| 2094 |
+
"learning_rate": 2.6048453738767755e-05,
|
| 2095 |
+
"loss": 0.6764,
|
| 2096 |
+
"step": 1480
|
| 2097 |
+
},
|
| 2098 |
+
{
|
| 2099 |
+
"epoch": 1.6615384615384614,
|
| 2100 |
+
"grad_norm": 0.07182079524548696,
|
| 2101 |
+
"learning_rate": 2.5228670125174704e-05,
|
| 2102 |
+
"loss": 0.6841,
|
| 2103 |
+
"step": 1485
|
| 2104 |
+
},
|
| 2105 |
+
{
|
| 2106 |
+
"epoch": 1.6671328671328671,
|
| 2107 |
+
"grad_norm": 0.072274928421158,
|
| 2108 |
+
"learning_rate": 2.4420807694358468e-05,
|
| 2109 |
+
"loss": 0.6823,
|
| 2110 |
+
"step": 1490
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 1.6727272727272728,
|
| 2114 |
+
"grad_norm": 0.07124500713293849,
|
| 2115 |
+
"learning_rate": 2.3624943632926853e-05,
|
| 2116 |
+
"loss": 0.6816,
|
| 2117 |
+
"step": 1495
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 1.6783216783216783,
|
| 2121 |
+
"grad_norm": 0.06940723884896695,
|
| 2122 |
+
"learning_rate": 2.2841153981112397e-05,
|
| 2123 |
+
"loss": 0.6805,
|
| 2124 |
+
"step": 1500
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 1.6839160839160838,
|
| 2128 |
+
"grad_norm": 0.07325019748524,
|
| 2129 |
+
"learning_rate": 2.20695136255073e-05,
|
| 2130 |
+
"loss": 0.6614,
|
| 2131 |
+
"step": 1505
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 1.6895104895104895,
|
| 2135 |
+
"grad_norm": 0.0727025067676513,
|
| 2136 |
+
"learning_rate": 2.1310096291908347e-05,
|
| 2137 |
+
"loss": 0.6851,
|
| 2138 |
+
"step": 1510
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 1.6951048951048953,
|
| 2142 |
+
"grad_norm": 0.07406758207149883,
|
| 2143 |
+
"learning_rate": 2.0562974538273024e-05,
|
| 2144 |
+
"loss": 0.6978,
|
| 2145 |
+
"step": 1515
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 1.7006993006993008,
|
| 2149 |
+
"grad_norm": 0.0684645890113912,
|
| 2150 |
+
"learning_rate": 1.9828219747786733e-05,
|
| 2151 |
+
"loss": 0.6814,
|
| 2152 |
+
"step": 1520
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 1.7062937062937062,
|
| 2156 |
+
"grad_norm": 0.07103130256598,
|
| 2157 |
+
"learning_rate": 1.910590212204281e-05,
|
| 2158 |
+
"loss": 0.6955,
|
| 2159 |
+
"step": 1525
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 1.7118881118881117,
|
| 2163 |
+
"grad_norm": 0.08660646823989741,
|
| 2164 |
+
"learning_rate": 1.839609067433495e-05,
|
| 2165 |
+
"loss": 0.6768,
|
| 2166 |
+
"step": 1530
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 1.7174825174825175,
|
| 2170 |
+
"grad_norm": 0.07000885068796428,
|
| 2171 |
+
"learning_rate": 1.7698853223063554e-05,
|
| 2172 |
+
"loss": 0.6814,
|
| 2173 |
+
"step": 1535
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 1.7230769230769232,
|
| 2177 |
+
"grad_norm": 0.07227717547773495,
|
| 2178 |
+
"learning_rate": 1.701425638525601e-05,
|
| 2179 |
+
"loss": 0.6863,
|
| 2180 |
+
"step": 1540
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 1.7286713286713287,
|
| 2184 |
+
"grad_norm": 0.08472297754432541,
|
| 2185 |
+
"learning_rate": 1.634236557020174e-05,
|
| 2186 |
+
"loss": 0.6739,
|
| 2187 |
+
"step": 1545
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 1.7342657342657342,
|
| 2191 |
+
"grad_norm": 0.07635635084022904,
|
| 2192 |
+
"learning_rate": 1.5683244973202848e-05,
|
| 2193 |
+
"loss": 0.6849,
|
| 2194 |
+
"step": 1550
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 1.7398601398601399,
|
| 2198 |
+
"grad_norm": 0.0733376716964744,
|
| 2199 |
+
"learning_rate": 1.5036957569440488e-05,
|
| 2200 |
+
"loss": 0.6736,
|
| 2201 |
+
"step": 1555
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 1.7454545454545456,
|
| 2205 |
+
"grad_norm": 0.06984335423469197,
|
| 2206 |
+
"learning_rate": 1.4403565107958142e-05,
|
| 2207 |
+
"loss": 0.6801,
|
| 2208 |
+
"step": 1560
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 1.751048951048951,
|
| 2212 |
+
"grad_norm": 0.07299591046039727,
|
| 2213 |
+
"learning_rate": 1.3783128105761649e-05,
|
| 2214 |
+
"loss": 0.6703,
|
| 2215 |
+
"step": 1565
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 1.7566433566433566,
|
| 2219 |
+
"grad_norm": 0.06856518432971286,
|
| 2220 |
+
"learning_rate": 1.3175705842037332e-05,
|
| 2221 |
+
"loss": 0.6811,
|
| 2222 |
+
"step": 1570
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 1.762237762237762,
|
| 2226 |
+
"grad_norm": 0.06892528407637724,
|
| 2227 |
+
"learning_rate": 1.2581356352488003e-05,
|
| 2228 |
+
"loss": 0.6895,
|
| 2229 |
+
"step": 1575
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 1.7678321678321678,
|
| 2233 |
+
"grad_norm": 0.0731333956161155,
|
| 2234 |
+
"learning_rate": 1.2000136423788226e-05,
|
| 2235 |
+
"loss": 0.6947,
|
| 2236 |
+
"step": 1580
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 1.7734265734265735,
|
| 2240 |
+
"grad_norm": 0.07510563722528918,
|
| 2241 |
+
"learning_rate": 1.1432101588158487e-05,
|
| 2242 |
+
"loss": 0.6782,
|
| 2243 |
+
"step": 1585
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 1.779020979020979,
|
| 2247 |
+
"grad_norm": 0.07335149840877105,
|
| 2248 |
+
"learning_rate": 1.0877306118059498e-05,
|
| 2249 |
+
"loss": 0.6832,
|
| 2250 |
+
"step": 1590
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 1.7846153846153845,
|
| 2254 |
+
"grad_norm": 0.07016146300305476,
|
| 2255 |
+
"learning_rate": 1.0335803021006783e-05,
|
| 2256 |
+
"loss": 0.6735,
|
| 2257 |
+
"step": 1595
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 1.7902097902097902,
|
| 2261 |
+
"grad_norm": 0.07110506905460164,
|
| 2262 |
+
"learning_rate": 9.807644034506024e-06,
|
| 2263 |
+
"loss": 0.6933,
|
| 2264 |
+
"step": 1600
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 1.795804195804196,
|
| 2268 |
+
"grad_norm": 0.07366692101954259,
|
| 2269 |
+
"learning_rate": 9.292879621110022e-06,
|
| 2270 |
+
"loss": 0.6775,
|
| 2271 |
+
"step": 1605
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 1.8013986013986014,
|
| 2275 |
+
"grad_norm": 0.07086756664676215,
|
| 2276 |
+
"learning_rate": 8.791558963597045e-06,
|
| 2277 |
+
"loss": 0.6847,
|
| 2278 |
+
"step": 1610
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 1.806993006993007,
|
| 2282 |
+
"grad_norm": 0.0704179943193231,
|
| 2283 |
+
"learning_rate": 8.30372996027195e-06,
|
| 2284 |
+
"loss": 0.6802,
|
| 2285 |
+
"step": 1615
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 1.8125874125874126,
|
| 2289 |
+
"grad_norm": 0.07178710401033903,
|
| 2290 |
+
"learning_rate": 7.829439220389521e-06,
|
| 2291 |
+
"loss": 0.6892,
|
| 2292 |
+
"step": 1620
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 1.8181818181818183,
|
| 2296 |
+
"grad_norm": 0.07500788961567526,
|
| 2297 |
+
"learning_rate": 7.368732059701499e-06,
|
| 2298 |
+
"loss": 0.6822,
|
| 2299 |
+
"step": 1625
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 1.8237762237762238,
|
| 2303 |
+
"grad_norm": 0.07030773912011665,
|
| 2304 |
+
"learning_rate": 6.921652496126623e-06,
|
| 2305 |
+
"loss": 0.6749,
|
| 2306 |
+
"step": 1630
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 1.8293706293706293,
|
| 2310 |
+
"grad_norm": 0.07046870723595076,
|
| 2311 |
+
"learning_rate": 6.4882432455452606e-06,
|
| 2312 |
+
"loss": 0.6748,
|
| 2313 |
+
"step": 1635
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 1.8349650349650348,
|
| 2317 |
+
"grad_norm": 0.07117585543308619,
|
| 2318 |
+
"learning_rate": 6.068545717717916e-06,
|
| 2319 |
+
"loss": 0.6828,
|
| 2320 |
+
"step": 1640
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 1.8405594405594405,
|
| 2324 |
+
"grad_norm": 0.06915527262243856,
|
| 2325 |
+
"learning_rate": 5.662600012328944e-06,
|
| 2326 |
+
"loss": 0.6883,
|
| 2327 |
+
"step": 1645
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 1.8461538461538463,
|
| 2331 |
+
"grad_norm": 0.06677186359807996,
|
| 2332 |
+
"learning_rate": 5.27044491515512e-06,
|
| 2333 |
+
"loss": 0.6701,
|
| 2334 |
+
"step": 1650
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 1.8517482517482518,
|
| 2338 |
+
"grad_norm": 0.06975135811394965,
|
| 2339 |
+
"learning_rate": 4.892117894359981e-06,
|
| 2340 |
+
"loss": 0.6896,
|
| 2341 |
+
"step": 1655
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 1.8573426573426572,
|
| 2345 |
+
"grad_norm": 0.06646442932877118,
|
| 2346 |
+
"learning_rate": 4.527655096913913e-06,
|
| 2347 |
+
"loss": 0.6736,
|
| 2348 |
+
"step": 1660
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 1.862937062937063,
|
| 2352 |
+
"grad_norm": 0.0696338374277028,
|
| 2353 |
+
"learning_rate": 4.177091345140488e-06,
|
| 2354 |
+
"loss": 0.6824,
|
| 2355 |
+
"step": 1665
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 1.8685314685314687,
|
| 2359 |
+
"grad_norm": 0.06913237278308428,
|
| 2360 |
+
"learning_rate": 3.840460133389434e-06,
|
| 2361 |
+
"loss": 0.6708,
|
| 2362 |
+
"step": 1670
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 1.8741258741258742,
|
| 2366 |
+
"grad_norm": 0.07472788742872244,
|
| 2367 |
+
"learning_rate": 3.5177936248364236e-06,
|
| 2368 |
+
"loss": 0.6843,
|
| 2369 |
+
"step": 1675
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 1.8797202797202797,
|
| 2373 |
+
"grad_norm": 0.07149423452101296,
|
| 2374 |
+
"learning_rate": 3.2091226484101506e-06,
|
| 2375 |
+
"loss": 0.6716,
|
| 2376 |
+
"step": 1680
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 1.8853146853146852,
|
| 2380 |
+
"grad_norm": 0.0676192443302248,
|
| 2381 |
+
"learning_rate": 2.9144766958466014e-06,
|
| 2382 |
+
"loss": 0.6816,
|
| 2383 |
+
"step": 1685
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 1.8909090909090909,
|
| 2387 |
+
"grad_norm": 0.07125330690039695,
|
| 2388 |
+
"learning_rate": 2.6338839188715433e-06,
|
| 2389 |
+
"loss": 0.686,
|
| 2390 |
+
"step": 1690
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 1.8965034965034966,
|
| 2394 |
+
"grad_norm": 0.07150788853257964,
|
| 2395 |
+
"learning_rate": 2.3673711265105754e-06,
|
| 2396 |
+
"loss": 0.6845,
|
| 2397 |
+
"step": 1695
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 1.902097902097902,
|
| 2401 |
+
"grad_norm": 0.06921602353027008,
|
| 2402 |
+
"learning_rate": 2.1149637825277953e-06,
|
| 2403 |
+
"loss": 0.6851,
|
| 2404 |
+
"step": 1700
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 1.9076923076923076,
|
| 2408 |
+
"grad_norm": 0.07103009932661779,
|
| 2409 |
+
"learning_rate": 1.876686002992861e-06,
|
| 2410 |
+
"loss": 0.6879,
|
| 2411 |
+
"step": 1705
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 1.9132867132867133,
|
| 2415 |
+
"grad_norm": 0.07078839389394587,
|
| 2416 |
+
"learning_rate": 1.6525605539768173e-06,
|
| 2417 |
+
"loss": 0.6842,
|
| 2418 |
+
"step": 1710
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 1.918881118881119,
|
| 2422 |
+
"grad_norm": 0.0677463790575426,
|
| 2423 |
+
"learning_rate": 1.4426088493769695e-06,
|
| 2424 |
+
"loss": 0.6822,
|
| 2425 |
+
"step": 1715
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 1.9244755244755245,
|
| 2429 |
+
"grad_norm": 0.0689426207416468,
|
| 2430 |
+
"learning_rate": 1.2468509488708534e-06,
|
| 2431 |
+
"loss": 0.671,
|
| 2432 |
+
"step": 1720
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 1.93006993006993,
|
| 2436 |
+
"grad_norm": 0.06820108494343277,
|
| 2437 |
+
"learning_rate": 1.0653055559997014e-06,
|
| 2438 |
+
"loss": 0.6775,
|
| 2439 |
+
"step": 1725
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 1.9356643356643357,
|
| 2443 |
+
"grad_norm": 0.06821597918112166,
|
| 2444 |
+
"learning_rate": 8.979900163813891e-07,
|
| 2445 |
+
"loss": 0.6701,
|
| 2446 |
+
"step": 1730
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 1.9412587412587412,
|
| 2450 |
+
"grad_norm": 0.06837657131866866,
|
| 2451 |
+
"learning_rate": 7.449203160532102e-07,
|
| 2452 |
+
"loss": 0.6725,
|
| 2453 |
+
"step": 1735
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 1.946853146853147,
|
| 2457 |
+
"grad_norm": 0.07000431064997485,
|
| 2458 |
+
"learning_rate": 6.061110799443991e-07,
|
| 2459 |
+
"loss": 0.6709,
|
| 2460 |
+
"step": 1740
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 1.9524475524475524,
|
| 2464 |
+
"grad_norm": 0.07108137280182684,
|
| 2465 |
+
"learning_rate": 4.815755704789481e-07,
|
| 2466 |
+
"loss": 0.6846,
|
| 2467 |
+
"step": 1745
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 1.958041958041958,
|
| 2471 |
+
"grad_norm": 0.07117415645327746,
|
| 2472 |
+
"learning_rate": 3.7132568630833804e-07,
|
| 2473 |
+
"loss": 0.6865,
|
| 2474 |
+
"step": 1750
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 1.9636363636363636,
|
| 2478 |
+
"grad_norm": 0.06606487210968705,
|
| 2479 |
+
"learning_rate": 2.753719611747474e-07,
|
| 2480 |
+
"loss": 0.6736,
|
| 2481 |
+
"step": 1755
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 1.9692307692307693,
|
| 2485 |
+
"grad_norm": 0.06965546022369541,
|
| 2486 |
+
"learning_rate": 1.9372356290460744e-07,
|
| 2487 |
+
"loss": 0.7023,
|
| 2488 |
+
"step": 1760
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 1.9748251748251748,
|
| 2492 |
+
"grad_norm": 0.06925022116566176,
|
| 2493 |
+
"learning_rate": 1.2638829253265316e-07,
|
| 2494 |
+
"loss": 0.6665,
|
| 2495 |
+
"step": 1765
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 1.9804195804195803,
|
| 2499 |
+
"grad_norm": 0.06743940727744605,
|
| 2500 |
+
"learning_rate": 7.337258355660236e-08,
|
| 2501 |
+
"loss": 0.6704,
|
| 2502 |
+
"step": 1770
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 1.986013986013986,
|
| 2506 |
+
"grad_norm": 0.06697392386506328,
|
| 2507 |
+
"learning_rate": 3.4681501322464386e-08,
|
| 2508 |
+
"loss": 0.6703,
|
| 2509 |
+
"step": 1775
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 1.9916083916083918,
|
| 2513 |
+
"grad_norm": 0.06864360169538268,
|
| 2514 |
+
"learning_rate": 1.0318742540560421e-08,
|
| 2515 |
+
"loss": 0.6753,
|
| 2516 |
+
"step": 1780
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 1.9972027972027973,
|
| 2520 |
+
"grad_norm": 0.06821105929452137,
|
| 2521 |
+
"learning_rate": 2.8663493232272684e-10,
|
| 2522 |
+
"loss": 0.6705,
|
| 2523 |
+
"step": 1785
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 1.9983216783216782,
|
| 2527 |
+
"eval_loss": 1.0629972219467163,
|
| 2528 |
+
"eval_runtime": 366.656,
|
| 2529 |
+
"eval_samples_per_second": 56.153,
|
| 2530 |
+
"eval_steps_per_second": 1.756,
|
| 2531 |
+
"step": 1786
|
| 2532 |
+
},
|
| 2533 |
+
{
|
| 2534 |
+
"epoch": 1.9983216783216782,
|
| 2535 |
+
"step": 1786,
|
| 2536 |
+
"total_flos": 1127451463778304.0,
|
| 2537 |
+
"train_loss": 0.777597175032935,
|
| 2538 |
+
"train_runtime": 16051.5455,
|
| 2539 |
+
"train_samples_per_second": 14.253,
|
| 2540 |
+
"train_steps_per_second": 0.111
|
| 2541 |
+
}
|
| 2542 |
+
],
|
| 2543 |
+
"logging_steps": 5,
|
| 2544 |
+
"max_steps": 1786,
|
| 2545 |
+
"num_input_tokens_seen": 0,
|
| 2546 |
+
"num_train_epochs": 2,
|
| 2547 |
+
"save_steps": 500,
|
| 2548 |
+
"stateful_callbacks": {
|
| 2549 |
+
"TrainerControl": {
|
| 2550 |
+
"args": {
|
| 2551 |
+
"should_epoch_stop": false,
|
| 2552 |
+
"should_evaluate": false,
|
| 2553 |
+
"should_log": false,
|
| 2554 |
+
"should_save": false,
|
| 2555 |
+
"should_training_stop": false
|
| 2556 |
+
},
|
| 2557 |
+
"attributes": {}
|
| 2558 |
+
}
|
| 2559 |
+
},
|
| 2560 |
+
"total_flos": 1127451463778304.0,
|
| 2561 |
+
"train_batch_size": 4,
|
| 2562 |
+
"trial_name": null,
|
| 2563 |
+
"trial_params": null
|
| 2564 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:290d7ecc3e2f19388b5c15b1a5cf555114883a038f5dc255f5173455bbbea8a2
|
| 3 |
+
size 6392
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|