metadata

tags:
  - sentence-transformers
  - sentence-similarity
  - feature-extraction
  - generated_from_trainer
  - dataset_size:50000
  - loss:CosineSimilarityLoss
base_model: google-bert/bert-base-uncased
widget:
  - source_sentence: >-
      Sometimes the people who represent themselves don't even know the
      significant facts of their case.
    sentences:
      - >-
        The law is very easy to understand, so representing yourself in court is
        the best way to win a case.
      - >-
        Sewage poured into upstairs windows from the streets while people
        whispered to each other.
      - His faith may be lacking.
  - source_sentence: >-
      When he married in 1901, he and his wife (Olga Knipper of the Moscow Art
      Theater) went directly from the ceremony to a honeymoon in a sanitarium.
    sentences:
      - if a person wants to eat you understand that
      - 'His wife has never went to a sanitarium. '
      - >-
        The new system appears far more complex, but ultimately easier and more
        thorough.
  - source_sentence: >-
      it really is i heard something that their supposed to be starting a huge
      campaign in New York about um child abuse and stopping child abuse and
      it's supposed to be like it's starting there supposed to be like a big
      nationwide campaign and you know so hopefully that will take off and
      really do something i don't know there's just
    sentences:
      - >-
        The Washington Post was the first company to report on attempts of
        private companies growing embryos.
      - Me too?
      - It's unfortunate that nobody is organizing a child abuse campaign.
  - source_sentence: >-
      On the mainland, an invasion of even greater significance followed in
      1580, when Philip II of Spain proclaimed himself king of Portugal and
      marched his armies across the border.
    sentences:
      - >-
        Some of the modern buildings that were erected in their place are not
        admired today.
      - Jon wanted to save them from the angry mob.
      - Philip II of Spain invaded Portugal.
  - source_sentence: The river plays a central role in all visits to Paris.
    sentences:
      - He said Dave Hanson.
      - The river is central to all vacations to Paris.
      - Trauma is the leading cause of alcohol abuse.
pipeline_tag: sentence-similarity
library_name: sentence-transformers
metrics:
  - pearson_cosine
  - spearman_cosine
model-index:
  - name: SentenceTransformer based on google-bert/bert-base-uncased
    results:
      - task:
          type: semantic-similarity
          name: Semantic Similarity
        dataset:
          name: Unknown
          type: unknown
        metrics:
          - type: pearson_cosine
            value: 0.7301988757371918
            name: Pearson Cosine
          - type: spearman_cosine
            value: 0.7323168725786805
            name: Spearman Cosine

SentenceTransformer based on google-bert/bert-base-uncased

This is a sentence-transformers model finetuned from google-bert/bert-base-uncased. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: google-bert/bert-base-uncased
Maximum Sequence Length: 512 tokens
Output Dimensionality: 768 dimensions
Similarity Function: Cosine Similarity

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("ryanhoangt/bert-base-uncased-mnli-cosine")
# Run inference
sentences = [
    'The river plays a central role in all visits to Paris.',
    'The river is central to all vacations to Paris.',
    'Trauma is the leading cause of alcohol abuse.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Semantic Similarity

Evaluated with EmbeddingSimilarityEvaluator

Metric	Value
pearson_cosine	0.7302
spearman_cosine	0.7323

Training Details

Training Dataset

Unnamed Dataset

Size: 50,000 training samples
Columns: sentence1, sentence2, and label
Approximate statistics based on the first 1000 samples:
sentence1 sentence2 label
type string string float
details
min: 4 tokens
mean: 26.95 tokens
max: 189 tokens

min: 5 tokens
mean: 14.11 tokens
max: 49 tokens

min: 0.0
mean: 0.34
max: 1.0

	sentence1	sentence2	label
type	string	string	float
details	min: 4 tokens mean: 26.95 tokens max: 189 tokens	min: 5 tokens mean: 14.11 tokens max: 49 tokens	min: 0.0 mean: 0.34 max: 1.0

Samples:

sentence1	sentence2	label
`Conceptually cream skimming has two basic dimensions - product and geography.`	`Product and geography are what make cream skimming work.`	`0.0`
`you know during the season and i guess at at your level uh you lose them to the next level if if they decide to recall the the parent team the Braves decide to call to recall a guy from triple A then a double A guy goes up to replace him and a single A guy goes up to replace him`	`You lose the things to the following level if the people recall.`	`1.0`
`One of our number will carry out your instructions minutely.`	`A member of my team will execute your orders with immense precision.`	`1.0`

Loss: CosineSimilarityLoss with these parameters:

{
    "loss_fct": "torch.nn.modules.loss.MSELoss"
}

Training Hyperparameters

Non-Default Hyperparameters

per_device_train_batch_size: 32
per_device_eval_batch_size: 32
num_train_epochs: 1
warmup_steps: 100
fp16: True

All Hyperparameters

Click to expand

overwrite_output_dir: False
do_predict: False
eval_strategy: no
prediction_loss_only: True
per_device_train_batch_size: 32
per_device_eval_batch_size: 32
per_gpu_train_batch_size: None
per_gpu_eval_batch_size: None
gradient_accumulation_steps: 1
eval_accumulation_steps: None
torch_empty_cache_steps: None
learning_rate: 5e-05
weight_decay: 0.0
adam_beta1: 0.9
adam_beta2: 0.999
adam_epsilon: 1e-08
max_grad_norm: 1.0
num_train_epochs: 1
max_steps: -1
lr_scheduler_type: linear
lr_scheduler_kwargs: {}
warmup_ratio: 0.0
warmup_steps: 100
log_level: passive
log_level_replica: warning
log_on_each_node: True
logging_nan_inf_filter: True
save_safetensors: True
save_on_each_node: False
save_only_model: False
restore_callback_states_from_checkpoint: False
no_cuda: False
use_cpu: False
use_mps_device: False
seed: 42
data_seed: None
jit_mode_eval: False
use_ipex: False
bf16: False
fp16: True
fp16_opt_level: O1
half_precision_backend: auto
bf16_full_eval: False
fp16_full_eval: False
tf32: None
local_rank: 0
ddp_backend: None
tpu_num_cores: None
tpu_metrics_debug: False
debug: []
dataloader_drop_last: False
dataloader_num_workers: 0
dataloader_prefetch_factor: None
past_index: -1
disable_tqdm: False
remove_unused_columns: True
label_names: None
load_best_model_at_end: False
ignore_data_skip: False
fsdp: []
fsdp_min_num_params: 0
fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}
fsdp_transformer_layer_cls_to_wrap: None
accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
deepspeed: None
label_smoothing_factor: 0.0
optim: adamw_torch
optim_args: None
adafactor: False
group_by_length: False
length_column_name: length
ddp_find_unused_parameters: None
ddp_bucket_cap_mb: None
ddp_broadcast_buffers: False
dataloader_pin_memory: True
dataloader_persistent_workers: False
skip_memory_metrics: True
use_legacy_prediction_loop: False
push_to_hub: False
resume_from_checkpoint: None
hub_model_id: None
hub_strategy: every_save
hub_private_repo: None
hub_always_push: False
gradient_checkpointing: False
gradient_checkpointing_kwargs: None
include_inputs_for_metrics: False
include_for_metrics: []
eval_do_concat_batches: True
fp16_backend: auto
push_to_hub_model_id: None
push_to_hub_organization: None
mp_parameters:
auto_find_batch_size: False
full_determinism: False
torchdynamo: None
ray_scope: last
ddp_timeout: 1800
torch_compile: False
torch_compile_backend: None
torch_compile_mode: None
include_tokens_per_second: False
include_num_input_tokens_seen: False
neftune_noise_alpha: None
optim_target_modules: None
batch_eval_metrics: False
eval_on_start: False
use_liger_kernel: False
eval_use_gather_object: False
average_tokens_across_devices: False
prompts: None
batch_sampler: batch_sampler
multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	spearman_cosine
0.0320	50	0.2752	-
0.0640	100	0.1898	-
0.0960	150	0.1733	-
0.1280	200	0.1679	-
0.1599	250	0.1743	-
0.1919	300	0.1703	-
0.2239	350	0.1599	-
0.2559	400	0.1614	-
0.2879	450	0.149	-
0.3199	500	0.1555	-
0.3519	550	0.1631	-
0.3839	600	0.1537	-
0.4159	650	0.1497	-
0.4479	700	0.1512	-
0.4798	750	0.157	-
0.5118	800	0.1544	-
0.5438	850	0.1502	-
0.5758	900	0.1459	-
0.6078	950	0.1476	-
0.6398	1000	0.1439	-
0.6718	1050	0.1508	-
0.7038	1100	0.1444	-
0.7358	1150	0.1457	-
0.7678	1200	0.1486	-
0.7997	1250	0.1485	-
0.8317	1300	0.1419	-
0.8637	1350	0.1406	-
0.8957	1400	0.1407	-
0.9277	1450	0.1434	-
0.9597	1500	0.1365	-
0.9917	1550	0.1465	-
-1	-1	-	0.7323

Framework Versions

Python: 3.11.12
Sentence Transformers: 4.1.0
Transformers: 4.52.2
PyTorch: 2.6.0+cu124
Accelerate: 1.7.0
Datasets: 3.2.0
Tokenizers: 0.21.1

Citation

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}