swardiantara/bert-tiny-snli-k5-fixed-euclidean

SentenceTransformer based on google/bert_uncased_L-2_H-128_A-2

This is a sentence-transformers model finetuned from google/bert_uncased_L-2_H-128_A-2. It maps sentences & paragraphs to a 128-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

• Model Type: Sentence Transformer
• Base model: google/bert_uncased_L-2_H-128_A-2
• Maximum Sequence Length: 128 tokens
• Output Dimensionality: 128 dimensions
• Similarity Function: Cosine Similarity
• Supported Modality: Text

Model Sources

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

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'transformer_task': 'feature-extraction', 'modality_config': {'text': {'method': 'forward', 'method_output_name': 'last_hidden_state'}}, 'module_output_name': 'token_embeddings', 'architecture': 'BertModel'})
  (1): Pooling({'embedding_dimension': 128, 'pooling_mode': 'mean', '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("swardiantara/bert-tiny-snli-k5-fixed-euclidean")
# Run inference
sentences = [
    'Children smiling and waving at camera [SEP] The kids are frowning',
    'A room full of girls raising their hands. [SEP] The boys are jumping on the trampoline.',
    'A woman is walking across the street eating a banana, while a man is following with his briefcase. [SEP] the woman is a seductress',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 128]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9991, 0.9962],
#         [0.9991, 1.0000, 0.9956],
#         [0.9962, 0.9956, 1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 6,043,097 training samples
• Columns: text_a, text_b, and label
• Approximate statistics based on the first 100 samples:

  	text_a	text_b	label
  type	string	string	list
  modality	text	text
  details	min: 13 tokens mean: 22.88 tokens max: 47 tokens	min: 20 tokens mean: 27.54 tokens max: 43 tokens	size: 2 elements

• Samples:

  text_a	text_b	label
  A person on a horse jumps over a broken down airplane. [SEP] A person is training his horse for a competition.	A woman is walking across the street eating a banana, while a man is following with his briefcase. [SEP] the woman is a seductress	[1.0, 0.0]
  A person on a horse jumps over a broken down airplane. [SEP] A person is training his horse for a competition.	A man on stilts is playing a tuba for money on the boardwalk. [SEP] A male street performer with a tuba is playing outside.	[0.0, 0.5]
  A person on a horse jumps over a broken down airplane. [SEP] A person is training his horse for a competition.	A boy is standing next to a car in front of a clothesline. [SEP] The boy is outside.	[0.0, 0.5]

• Loss: main.OrdinalProxyContrastiveLoss

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 1024
• learning_rate: 1e-05
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• per_device_train_batch_size: 1024
• num_train_epochs: 3
• max_steps: -1
• learning_rate: 1e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch
• optim_args: None
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• optim_target_modules: None
• gradient_accumulation_steps: 1
• average_tokens_across_devices: True
• max_grad_norm: 1.0
• label_smoothing_factor: 0.0
• bf16: False
• fp16: False
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• use_liger_kernel: False
• liger_kernel_config: None
• use_cache: False
• neftune_noise_alpha: None
• torch_empty_cache_steps: None
• auto_find_batch_size: False
• log_on_each_node: True
• logging_nan_inf_filter: True
• include_num_input_tokens_seen: no
• log_level: passive
• log_level_replica: warning
• disable_tqdm: False
• project: huggingface
• trackio_space_id: None
• trackio_bucket_id: None
• trackio_static_space_id: None
• per_device_eval_batch_size: 8
• prediction_loss_only: True
• eval_on_start: False
• eval_do_concat_batches: True
• eval_use_gather_object: False
• eval_accumulation_steps: None
• include_for_metrics: []
• batch_eval_metrics: False
• save_only_model: False
• save_on_each_node: False
• enable_jit_checkpoint: False
• push_to_hub: False
• hub_private_repo: None
• hub_model_id: None
• hub_strategy: every_save
• hub_always_push: False
• hub_revision: None
• load_best_model_at_end: True
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}
• parallelism_config: None
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_pin_memory: True
• dataloader_persistent_workers: False
• dataloader_prefetch_factor: None
• remove_unused_columns: True
• label_names: None
• train_sampling_strategy: random
• length_column_name: length
• ddp_find_unused_parameters: None
• ddp_bucket_cap_mb: None
• ddp_broadcast_buffers: False
• ddp_static_graph: None
• ddp_backend: None
• ddp_timeout: 1800
• fsdp: None
• fsdp_config: None
• deepspeed: None
• debug: []
• skip_memory_metrics: True
• do_predict: False
• resume_from_checkpoint: None
• warmup_ratio: None
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

Epoch	Step	Training Loss
0.0847	500	0.4598
0.1694	1000	0.0829
0.2542	1500	0.0778
0.3389	2000	0.0752
0.4236	2500	0.0734
0.5083	3000	0.0716
0.5930	3500	0.0712
0.6777	4000	0.0704
0.7625	4500	0.0702
0.8472	5000	0.0700
0.9319	5500	0.0691
1.0	5902	-
1.0166	6000	0.0694
1.1013	6500	0.0690
1.1860	7000	0.0688
1.2708	7500	0.0683
1.3555	8000	0.0687
1.4402	8500	0.0682
1.5249	9000	0.0681
1.6096	9500	0.0679
1.6943	10000	0.0681
1.7791	10500	0.0684
1.8638	11000	0.0677
1.9485	11500	0.0675
2.0	11804	-
2.0332	12000	0.0674
2.1179	12500	0.0671
2.2026	13000	0.0668
2.2874	13500	0.0669
2.3721	14000	0.0660
2.4568	14500	0.0666
2.5415	15000	0.0663
2.6262	15500	0.0665
2.7109	16000	0.0660
2.7957	16500	0.0666
2.8804	17000	0.0660
2.9651	17500	0.0658
3.0	17706	-

• The bold row denotes the saved checkpoint.

Training Time

• Training: 36.2 minutes
• Evaluation: 2.6 seconds
• Total: 36.3 minutes

Framework Versions

• Python: 3.12.4
• Sentence Transformers: 5.5.1
• Transformers: 5.11.0
• PyTorch: 2.5.1+cu121
• Accelerate: 1.13.0
• Datasets: 2.21.0
• Tokenizers: 0.22.2

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",
}