SentenceTransformer based on VinitT/Embeddings-Trivia

This is a sentence-transformers model finetuned from VinitT/Embeddings-Trivia on the all-nli dataset. It maps sentences & paragraphs to a 384-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: VinitT/Embeddings-Trivia
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • all-nli
• Language: en

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, 'architecture': 'BertModel'})
  (1): Pooling({'word_embedding_dimension': 384, '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})
  (2): Normalize()
)

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("sentence_transformers_model_id")
# Run inference
sentences = [
    'so he has overcome alcoholism at this point',
    "He's gotten stronger and has overcome alcoholism.",
    "He still is a heavy drinker and can't control it.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.7603, 0.0849],
#         [0.7603, 1.0000, 0.0794],
#         [0.0849, 0.0794, 1.0000]])

Evaluation

Metrics

Triplet

• Dataset: contra_eval
• Evaluated with TripletEvaluator

Training Details

Training Dataset

all-nli

• Dataset: all-nli at d482672
• Size: 556,850 training samples
• Columns: anchor, positive, negative, and label
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	label
  type	string	string	string	int
  details	min: 5 tokensmean: 19.16 tokensmax: 194 tokens	min: 5 tokensmean: 11.86 tokensmax: 32 tokens	min: 5 tokensmean: 12.23 tokensmax: 37 tokens	1: 100.00%

• Samples:

  anchor	positive	negative	label
  a young girl wearing blue smiles.	A little girl wears blue.	A little girl frowns as she wears an ugly burlap sack.	1
  An old man wearing a tan jacket and blue pants standing on a sidewalk with a small suitcase.	A man wearing a jacket and jeans holds a suitcase.	A young woman sits on a bench holding her purse.	1
  The people are inside.	Two people are dancing by a red couch.	People walk up and down the steps in front of a church.	1

• Loss: custom_loss.ContradictionMarginLoss with these parameters:

  {
      "margin_neutral": 0.2,
      "margin_contradiction": 0.4
  }
  

Evaluation Dataset

all-nli

• Dataset: all-nli at d482672
• Size: 1,000 evaluation samples
• Columns: anchor, positive, negative, and label
• Approximate statistics based on the first 1000 samples:

  	anchor	positive	negative	label
  type	string	string	string	int
  details	min: 5 tokensmean: 18.67 tokensmax: 86 tokens	min: 4 tokensmean: 11.92 tokensmax: 41 tokens	min: 4 tokensmean: 12.13 tokensmax: 40 tokens	1: 100.00%

• Samples:

  anchor	positive	negative	label
  An older man riding a bike.	An elderly man is biking	an old man is sleeping	1
  The man is on a skateboard.	A shirtless man is doing a skateboard trick over a bike rail.	A man performs a bike trick on a ramp.	1
  The Episcopalians are all going to hell.	The Episcopalians will not be going to heaven.	All Episcopalians will go to heaven.	1

• Loss: custom_loss.ContradictionMarginLoss with these parameters:

  {
      "margin_neutral": 0.2,
      "margin_contradiction": 0.4
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• learning_rate: 2e-05
• weight_decay: 0.01
• num_train_epochs: 1
• warmup_ratio: 0.1
• warmup_steps: 0.1
• fp16: True
• load_best_model_at_end: True

All Hyperparameters

• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 64
• per_device_eval_batch_size: 64
• gradient_accumulation_steps: 1
• eval_accumulation_steps: None
• torch_empty_cache_steps: None
• learning_rate: 2e-05
• weight_decay: 0.01
• 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: None
• warmup_ratio: 0.1
• warmup_steps: 0.1
• log_level: passive
• log_level_replica: warning
• log_on_each_node: True
• logging_nan_inf_filter: True
• enable_jit_checkpoint: False
• save_on_each_node: False
• save_only_model: False
• restore_callback_states_from_checkpoint: False
• use_cpu: False
• seed: 42
• data_seed: None
• bf16: False
• fp16: True
• bf16_full_eval: False
• fp16_full_eval: False
• tf32: None
• local_rank: -1
• ddp_backend: None
• debug: []
• dataloader_drop_last: False
• dataloader_num_workers: 0
• dataloader_prefetch_factor: None
• disable_tqdm: False
• remove_unused_columns: True
• label_names: None
• load_best_model_at_end: True
• ignore_data_skip: False
• fsdp: []
• fsdp_config: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': 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
• deepspeed: None
• label_smoothing_factor: 0.0
• optim: adamw_torch_fused
• optim_args: None
• group_by_length: False
• length_column_name: length
• project: huggingface
• trackio_space_id: trackio
• 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
• push_to_hub: False
• resume_from_checkpoint: None
• hub_model_id: None
• hub_strategy: every_save
• hub_private_repo: None
• hub_always_push: False
• hub_revision: None
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_for_metrics: []
• eval_do_concat_batches: True
• auto_find_batch_size: False
• full_determinism: False
• ddp_timeout: 1800
• torch_compile: False
• torch_compile_backend: None
• torch_compile_mode: None
• include_num_input_tokens_seen: no
• neftune_noise_alpha: None
• optim_target_modules: None
• batch_eval_metrics: False
• eval_on_start: False
• use_liger_kernel: False
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: True
• use_cache: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

Training Logs

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.12.12
• Sentence Transformers: 5.2.2
• Transformers: 5.0.0
• PyTorch: 2.9.0+cu128
• Accelerate: 1.12.0
• Datasets: 4.0.0
• 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",
}