gabrielegabellone/bert-base-uncased-snli

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: 256 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': 256, '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("gabrielegabellone/bert-base-uncased-snli")
# Run inference
sentences = [
    'A dog running through tall grass.',
    'Dog lying in an open field.',
    'The woman is outside.',
]
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]

Training Details

Training Dataset

Unnamed Dataset

• Size: 549,367 training samples
• Columns: sentence_0, sentence_1, and label
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	int
  details	min: 6 tokens mean: 16.75 tokens max: 73 tokens	min: 4 tokens mean: 10.57 tokens max: 29 tokens	0: ~36.60% 1: ~33.00% 2: ~30.40%

• Samples:

  sentence_0	sentence_1	label
  A woman in wearing a white hat holding a scythe and a cutting of wheat.	A sad woman in wearing a white hat holding a scythe and a cutting of wheat.	1
  There is an old man sitting on the edge of a concrete pillar wearing a hat.	A man sits outside.	0
  A man dressed as a cook enjoys a meal and bottle of wine in an empty restaurant.	There is a chef that is waiting for people to come to his restaurant and is preparing food.	1

• Loss: SoftmaxLoss

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 1
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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
• num_train_epochs: 1
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• warmup_steps: 0
• 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: False
• 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: round_robin

Training Logs

Epoch	Step	Training Loss
0.0146	500	0.9507
0.0291	1000	0.8171
0.0437	1500	0.7619
0.0582	2000	0.7505
0.0728	2500	0.7221
0.0874	3000	0.6913
0.1019	3500	0.6753
0.1165	4000	0.6658
0.1311	4500	0.6578
0.1456	5000	0.6438
0.1602	5500	0.6348
0.1747	6000	0.6286
0.1893	6500	0.6261
0.2039	7000	0.6154
0.2184	7500	0.6184
0.2330	8000	0.6107
0.2476	8500	0.6189
0.2621	9000	0.6087
0.2767	9500	0.612
0.2912	10000	0.6026
0.3058	10500	0.603
0.3204	11000	0.5987
0.3349	11500	0.5999
0.3495	12000	0.5893
0.3640	12500	0.5767
0.3786	13000	0.59
0.3932	13500	0.5898
0.4077	14000	0.5854
0.4223	14500	0.571
0.4369	15000	0.5811
0.4514	15500	0.569
0.4660	16000	0.5751
0.4805	16500	0.5714
0.4951	17000	0.5605
0.5097	17500	0.5687
0.5242	18000	0.5566
0.5388	18500	0.5642
0.5534	19000	0.5468
0.5679	19500	0.5733
0.5825	20000	0.5582
0.5970	20500	0.5534
0.6116	21000	0.5417
0.6262	21500	0.5483
0.6407	22000	0.5446
0.6553	22500	0.5581
0.6699	23000	0.5353
0.6844	23500	0.5513
0.6990	24000	0.5521
0.7135	24500	0.5402
0.7281	25000	0.5371
0.7427	25500	0.554
0.7572	26000	0.5322
0.7718	26500	0.5404
0.7863	27000	0.5372
0.8009	27500	0.5465
0.8155	28000	0.5387
0.8300	28500	0.5363
0.8446	29000	0.5296
0.8592	29500	0.5356
0.8737	30000	0.5341
0.8883	30500	0.54
0.9028	31000	0.527
0.9174	31500	0.5261
0.9320	32000	0.5424
0.9465	32500	0.5286
0.9611	33000	0.5262
0.9757	33500	0.5255
0.9902	34000	0.5221

Framework Versions

• Python: 3.11.12
• Sentence Transformers: 4.1.0
• Transformers: 4.52.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.7.0
• Datasets: 3.6.0
• Tokenizers: 0.21.1

Citation

BibTeX

Sentence Transformers and SoftmaxLoss

@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",
}