Jeswina-123/AI_QA_Complete_Project_Models

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

This is a sentence-transformers model finetuned from sentence-transformers/all-MiniLM-L6-v2. It maps sentences & paragraphs to a 384-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, 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
• 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': 384, 'pooling_mode': 'mean', '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 = [
    "What is the modern name for the West Saxon dialect of Alfred's time?",
    'Each of these four dialects was associated with an independent kingdom on the island. Of these, Northumbria south of the Tyne, and most of Mercia, were overrun by the Vikings during the 9th century. The portion of Mercia that was successfully defended, and all of Kent, were then integrated into Wessex under Alfred the Great. From that time on, the West Saxon dialect (then in the form now known as Early West Saxon) became standardised as the language of government, and as the basis for the many works of literature and religious materials produced or translated from Latin in that period.',
    "At the dawn of the Grand Lodge era, during the 1720s, James Anderson composed the first printed constitutions for Freemasons, the basis for most subsequent constitutions, which specifically excluded women from Freemasonry. As Freemasonry spread, continental masons began to include their ladies in Lodges of Adoption, which worked three degrees with the same names as the men's but different content. The French officially abandoned the experiment in the early 19th century. Later organisations with a similar aim emerged in the United States, but distinguished the names of the degrees from those of male masonry.",
]
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,  0.5936,  0.0054],
#         [ 0.5936,  1.0000, -0.0376],
#         [ 0.0054, -0.0376,  1.0000]])

Training Details

Training Dataset

Unnamed Dataset

• Size: 10,000 training samples
• Columns: sentence_0 and sentence_1
• Approximate statistics based on the first 100 samples:

  	sentence_0	sentence_1
  type	string	string
  modality	text	text
  details	min: 8 tokens mean: 14.67 tokens max: 30 tokens	min: 33 tokens mean: 152.05 tokens max: 256 tokens

• Samples:

  sentence_0	sentence_1
  Plato and Aristotle are known for systematically discussing what?	Subsequently, Plato and Aristotle produced the first systematic discussions of natural philosophy, which did much to shape later investigations of nature. Their development of deductive reasoning was of particular importance and usefulness to later scientific inquiry. Plato founded the Platonic Academy in 387 BC, whose motto was "Let none unversed in geometry enter here", and turned out many notable philosophers. Plato's student Aristotle introduced empiricism and the notion that universal truths can be arrived at via observation and induction, thereby laying the foundations of the scientific method. Aristotle also produced many biological writings that were empirical in nature, focusing on biological causation and the diversity of life. He made countless observations of nature, especially the habits and attributes of plants and animals in the world around him, classified more than 540 animal species, and dissected at least 50. Aristotle's writings profoundly influenced subsequent Isla...
  Destiny's Child released a compilation album about which topic?	In January 2013, Destiny's Child released Love Songs, a compilation album of the romance-themed songs from their previous albums and a newly recorded track, "Nuclear". Beyoncé performed the American national anthem singing along with a pre-recorded track at President Obama's second inauguration in Washington, D.C. The following month, Beyoncé performed at the Super Bowl XLVII halftime show, held at the Mercedes-Benz Superdome in New Orleans. The performance stands as the second most tweeted about moment in history at 268,000 tweets per minute. At the 55th Annual Grammy Awards, Beyoncé won for Best Traditional R&B Performance for "Love on Top". Her feature-length documentary film, Life Is But a Dream, first aired on HBO on February 16, 2013. The film, which she directed and produced herself, featured footage from her childhood, her as a mother and businesswoman, recording, rehearsing for live performances, and her return to the spotlight following Blue Ivy's birth. Its DVD release in No...
  Which church is located on East 90th Street between Second and Third Avenue?	90th Street is split into two segments. The first segment, West 90th Street begins at Riverside Drive and ends at Central Park West or West Drive, when it is open, in Central Park on the Upper West Side. The second segment of East 90th Street begins at East Drive, at Engineers Gate of Central Park. When East Drive is closed, East 90th Street begins at Fifth Avenue on the Upper East Side and curves to the right at the FDR Drive becoming East End Avenue. Our Lady of Good Counsel Church, is located on East 90th Street between Third Avenue and Second Avenue, across the street from Ruppert Towers (1601 and 1619 Third Avenue) and Ruppert Park. Asphalt Green, which is located on East 90th Street between York Avenue and East End Avenue.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 32
• per_device_eval_batch_size: 32
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

Click to expand

• per_device_train_batch_size: 32
• num_train_epochs: 3
• max_steps: -1
• learning_rate: 5e-05
• lr_scheduler_type: linear
• lr_scheduler_kwargs: None
• warmup_steps: 0
• optim: adamw_torch_fused
• 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
• 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: 32
• 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: False
• 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
• dataloader_multiprocessing_context: None
• dataloader_in_order: True
• 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
• local_rank: -1
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}
• warmup_ratio: None

Training Logs

Epoch	Step	Training Loss
1.5974	500	0.0709

Training Time

• Training: 12.2 hours

Framework Versions

• Python: 3.11.0
• Sentence Transformers: 5.7.0
• Transformers: 5.17.0
• PyTorch: 2.14.0+cpu
• Accelerate: 1.15.0
• Datasets: 4.8.5
• Tokenizers: 0.23.2

Additional Resources

• Training and Finetuning Embedding Models with Sentence Transformers: the end-to-end guide for training or finetuning Sentence Transformer models.
• Introduction to Matryoshka Embedding Models: variable-size embeddings that can be truncated with minimal quality loss.
• Binary and Scalar Embedding Quantization for Significantly Faster & Cheaper Retrieval: post-training compression of embedding vectors.
• Multimodal Embedding & Reranker Models with Sentence Transformers: use text, image, audio, and video models through the same API.
• Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers: train multimodal embedding models, with a Visual Document Retrieval walkthrough.

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",
}

MultipleNegativesRankingLoss

@misc{oord2019representationlearningcontrastivepredictive,
      title={Representation Learning with Contrastive Predictive Coding},
      author={Aaron van den Oord and Yazhe Li and Oriol Vinyals},
      year={2019},
      eprint={1807.03748},
      archivePrefix={arXiv},
      primaryClass={cs.LG},
      url={https://arxiv.org/abs/1807.03748},
}