Pyro-X2/distilbert-base-uncased-sts

SentenceTransformer based on distilbert/distilbert-base-uncased

This is a sentence-transformers model finetuned from distilbert/distilbert-base-uncased on the csv dataset. 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: distilbert/distilbert-base-uncased
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • csv

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: DistilBertModel 
  (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("Pyro-X2/distilbert-base-uncased-sts")
# Run inference
sentences = [
    'Tell us what the charges were.',
    'Yes what are his charges.',
    'A person is buttering a tray.',
]
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

• Datasets: sts-dev and sts-test
• Evaluated with EmbeddingSimilarityEvaluator

Metric	sts-dev	sts-test
pearson_cosine	0.378	0.349
spearman_cosine	0.4731	0.4491

Training Details

Training Dataset

csv

• Dataset: csv
• Size: 4,480 training samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 1000 samples:

  	sentence1	sentence2	score
  type	string	string	int
  details	min: 6 tokens mean: 15.14 tokens max: 50 tokens	min: 6 tokens mean: 15.07 tokens max: 52 tokens	0: ~14.20% 1: ~11.60% 2: ~18.40% 3: ~23.30% 4: ~21.70% 5: ~10.80%

• Samples:

  sentence1	sentence2	score
  A man is speaking.	A man is spitting.	1
  Austrian found hoarding 56 stolen skulls in home museum	Austrian man charged after 56 human skulls are found at his home	4
  Mitt Romney wins Republican primary in Indiana	Romney wins Florida Republican primary	2

• Loss: CosineSimilarityLoss with these parameters:

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

Evaluation Dataset

csv

• Dataset: csv
• Size: 560 evaluation samples
• Columns: sentence1, sentence2, and score
• Approximate statistics based on the first 560 samples:

  	sentence1	sentence2	score
  type	string	string	int
  details	min: 5 tokens mean: 14.41 tokens max: 44 tokens	min: 5 tokens mean: 14.28 tokens max: 42 tokens	0: ~12.86% 1: ~16.96% 2: ~14.82% 3: ~18.21% 4: ~26.43% 5: ~10.71%

• Samples:

  sentence1	sentence2	score
  An airplane is flying in the air.	A South African Airways plane is flying in a blue sky.	3
  A television, upholstered chair, and coffee stable in a bright room.	A leather couch and wooden table in a living room.	2
  Red panda’s short-lived zoo escape	India’s march to Mars	0

• Loss: CosineSimilarityLoss with these parameters:

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

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 4
• warmup_ratio: 0.1

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• 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: 1.0
• num_train_epochs: 4
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.1
• 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
• dispatch_batches: None
• split_batches: 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	Validation Loss	sts-dev_spearman_cosine	sts-test_spearman_cosine
0.3571	100	5.031	5.0990	0.4973	-
0.7143	200	4.9152	5.0985	0.4944	-
1.0714	300	4.8198	5.0984	0.4959	-
1.4286	400	4.9102	5.0983	0.4884	-
1.7857	500	4.9238	5.0983	0.4798	-
2.1429	600	4.9387	5.0983	0.4777	-
2.5	700	4.8955	5.0983	0.4752	-
2.8571	800	4.9623	5.0983	0.4740	-
3.2143	900	4.7754	5.0983	0.4739	-
3.5714	1000	4.936	5.0983	0.4734	-
3.9286	1100	4.9254	5.0983	0.4731	-
-1	-1	-	-	-	0.4491

Framework Versions

• Python: 3.12.12
• Sentence Transformers: 4.1.0
• Transformers: 4.49.0
• PyTorch: 2.3.0.post101
• Accelerate: 1.10.1
• Datasets: 3.3.2
• Tokenizers: 0.21.0

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",
}