manupande21/all-MiniLM-L6-v2-finetuned-triples_hard_negatives

SentenceTransformer based on sentence-transformers/all-MiniLM-L6-v2

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2 using the dataset : manupande21/msmarco_train_hard_negatives. 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: sentence-transformers/all-MiniLM-L6-v2
• Maximum Sequence Length: 256 tokens
• Output Dimensionality: 384 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': 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})
)

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("manupande21/all-MiniLM-L6-v2-finetuned-triples_hard_negatives")
# Run inference
sentences = [
    'viral meningitis contagious',
    'meningitis is contagious prolonged close contact can spread the bacteria that cause meningitis the bacteria can be spread through kissing coughs and sneezes shared cutlery or sharing items like toothbrushes or cigarettes',
    'Infectious refers to a disease involving a microorganism that can be transmitted from one person to another only by a specific kind of contact; venereal diseases are usually infectious. In nontechnical senses, contagious emphasizes the rapidity with which something spreads: Contagious laughter ran through the hall. Infectious suggests the pleasantly irresistible quality of something: Her infectious good humor made her a popular guest.',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 384]

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

Evaluation

Metrics

Triplet

• Dataset: test-eval
• Evaluated with TripletEvaluator

Metric	Value
cosine_accuracy	0.8426

Training Details

Training Dataset

Unnamed Dataset

• Size: 402,351 training samples
• Columns: sentence_0, sentence_1, and sentence_2
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	sentence_2
  type	string	string	string
  details	min: 4 tokens mean: 8.97 tokens max: 30 tokens	min: 21 tokens mean: 80.49 tokens max: 242 tokens	min: 18 tokens mean: 76.84 tokens max: 198 tokens

• Samples:

  sentence_0	sentence_1	sentence_2
  how much of an ira distribution is taxable	Instead of a distributionof $10,000, let’s say the IRA ownertakes out a distribution of $25,000.This pushes the MAGI above$44,000, making $17,000 of SocialSecurity taxable.However, you can control howmuch tax you pay on Social Securityin many instances by controllingyour IRA distribution strategy.	And regardless, whatever earnings you have on your contributions won't be taxed until you withdraw that money many years later. For example, let's say you made $30,000 during the year, and you put $2,000 of it into an IRA. You would pay income tax on only $28,000.
  who is archer fate ubl	Archer is a Shirou who is from a different universe and who faced different circumstances and got betrayed and jumped to the conclusion that Kiritsugu's ideals were nothing but bs and that world peace can never be achieved because conflict is a part of human nature.	Archer's True Name is Gilgamesh, the great half-god, half-human king born from the union between the King of Uruk, Lugalbanda, and goddess Rimat-Ninsun. He ruled the Sumerian city-state of Uruk, the capital city of ancient Mesopotamia in B.C. era.
  what is comvault	Commvault software is an enterprise data protection and information management suite built on a scalable, single platform and unifying code base.The product uses a common set of advanced capabilities related to the storage and access of data and are administered through one console application.ommvault software is an enterprise data protection and information management suite built on a scalable, single platform and unifying code base.	Definition of commensurate. 1 1 : equal in measure or extent : coextensive lived a life commensurate with the early years of the republic. 2 2 : corresponding in size, extent, amount, or degree : proportionate was given a job commensurate with her abilities. 3 3 : commensurable 1.

• Loss: TripletLoss with these parameters:

  {
      "distance_metric": "TripletDistanceMetric.EUCLIDEAN",
      "triplet_margin": 5
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: steps
• per_device_train_batch_size: 1024
• per_device_eval_batch_size: 1024
• num_train_epochs: 5
• fp16: True
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 1024
• per_device_eval_batch_size: 1024
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 1
• eval_accumulation_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: 5
• 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: 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: False
• hub_always_push: False
• gradient_checkpointing: False
• gradient_checkpointing_kwargs: None
• include_inputs_for_metrics: False
• 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
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin

Training Logs

Epoch	Step	Training Loss	test-eval_cosine_accuracy
1.0	393	-	0.8191
1.2723	500	3.0326	-
2.0	786	-	0.8353
2.5445	1000	2.4038	0.8376
3.0	1179	-	0.8426

Framework Versions

• Python: 3.11.5
• Sentence Transformers: 4.1.0
• Transformers: 4.41.0
• PyTorch: 2.7.0+cu126
• Accelerate: 1.7.0
• Datasets: 3.2.0
• Tokenizers: 0.19.1

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",
}

TripletLoss

@misc{hermans2017defense,
    title={In Defense of the Triplet Loss for Person Re-Identification},
    author={Alexander Hermans and Lucas Beyer and Bastian Leibe},
    year={2017},
    eprint={1703.07737},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}