MANMEET75/bge-large-Chatbot-matryoshka

BGE large Chatbot Matryoshka

This is a sentence-transformers model finetuned from BAAI/bge-large-en-v1.5. It maps sentences & paragraphs to a 1024-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: BAAI/bge-large-en-v1.5
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 tokens
• Similarity Function: Cosine Similarity
• Language: en
• License: apache-2.0

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': True}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 1024, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, '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("MANMEET75/bge-large-Chatbot-matryoshka")
# Run inference
sentences = [
    "I can understand and respond in multiple Indian regional languages. Feel free to communicate with me in the language you're most comfortable with.",
    'Bharti, what languages can you understand and respond to?',
    'Bharti, can you provide tips for effective online communication?',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 1024]

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

Evaluation

Metrics

Information Retrieval

• Dataset: dim_768
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8837
cosine_accuracy@3	0.9535
cosine_accuracy@5	0.9535
cosine_accuracy@10	0.9535
cosine_precision@1	0.8837
cosine_precision@3	0.3178
cosine_precision@5	0.1907
cosine_precision@10	0.0953
cosine_recall@1	0.8837
cosine_recall@3	0.9535
cosine_recall@5	0.9535
cosine_recall@10	0.9535
cosine_ndcg@10	0.9247
cosine_mrr@10	0.9147
cosine_map@100	0.9186

Information Retrieval

• Dataset: dim_512
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8837
cosine_accuracy@3	0.9535
cosine_accuracy@5	0.9535
cosine_accuracy@10	0.9535
cosine_precision@1	0.8837
cosine_precision@3	0.3178
cosine_precision@5	0.1907
cosine_precision@10	0.0953
cosine_recall@1	0.8837
cosine_recall@3	0.9535
cosine_recall@5	0.9535
cosine_recall@10	0.9535
cosine_ndcg@10	0.9247
cosine_mrr@10	0.9147
cosine_map@100	0.9186

Information Retrieval

• Dataset: dim_256
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8837
cosine_accuracy@3	0.9302
cosine_accuracy@5	0.9535
cosine_accuracy@10	0.9535
cosine_precision@1	0.8837
cosine_precision@3	0.3101
cosine_precision@5	0.1907
cosine_precision@10	0.0953
cosine_recall@1	0.8837
cosine_recall@3	0.9302
cosine_recall@5	0.9535
cosine_recall@10	0.9535
cosine_ndcg@10	0.9221
cosine_mrr@10	0.9116
cosine_map@100	0.9148

Information Retrieval

• Dataset: dim_128
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.907
cosine_accuracy@3	0.9302
cosine_accuracy@5	0.9302
cosine_accuracy@10	0.9535
cosine_precision@1	0.907
cosine_precision@3	0.3101
cosine_precision@5	0.186
cosine_precision@10	0.0953
cosine_recall@1	0.907
cosine_recall@3	0.9302
cosine_recall@5	0.9302
cosine_recall@10	0.9535
cosine_ndcg@10	0.9299
cosine_mrr@10	0.9225
cosine_map@100	0.9255

Information Retrieval

• Dataset: dim_64
• Evaluated with InformationRetrievalEvaluator

Metric	Value
cosine_accuracy@1	0.8605
cosine_accuracy@3	0.9535
cosine_accuracy@5	0.9767
cosine_accuracy@10	0.9767
cosine_precision@1	0.8605
cosine_precision@3	0.3178
cosine_precision@5	0.1953
cosine_precision@10	0.0977
cosine_recall@1	0.8605
cosine_recall@3	0.9535
cosine_recall@5	0.9767
cosine_recall@10	0.9767
cosine_ndcg@10	0.9261
cosine_mrr@10	0.9089
cosine_map@100	0.9089

Training Details

Training Dataset

Unnamed Dataset

• Size: 530 training samples
• Columns: positive and anchor
• Approximate statistics based on the first 1000 samples:

  	positive	anchor
  type	string	string
  details	min: 11 tokens mean: 35.33 tokens max: 99 tokens	min: 7 tokens mean: 17.3 tokens max: 29 tokens

• Samples:

  positive	anchor
  BharatPe Speaker comes with the following benefits: - Helps you avoid payment fraud - Lightweight & Easy installation process - Compatible with SIM & GPRS connectivity - Comes with a battery, no hassle of constant charging - Available in 10 Languages - Cashback Offers - Free replacement To Know more and place an order, tap below http://bharatpe.in/speaker.	What are the benefits of the BharatPe speaker?
  BharatPe Speaker comes with the following benefits: - Helps you avoid payment fraud - Lightweight & Easy installation process - Compatible with SIM & GPRS connectivity - Comes with a battery, no hassle of constant charging - Available in 10 Languages - Cashback Offers - Free replacement To Know more and place an order, tap below http://bharatpe.in/speaker.	What advantages does the BharatPe speaker offer?
  BharatPe Speaker comes with the following benefits: - Helps you avoid payment fraud - Lightweight & Easy installation process - Compatible with SIM & GPRS connectivity - Comes with a battery, no hassle of constant charging - Available in 10 Languages - Cashback Offers - Free replacement To Know more and place an order, tap below http://bharatpe.in/speaker.	Can you outline the benefits of using the BharatPe speaker?

• Loss: MatryoshkaLoss with these parameters:

  {
      "loss": "MultipleNegativesRankingLoss",
      "matryoshka_dims": [
          768,
          512,
          256,
          128,
          64
      ],
      "matryoshka_weights": [
          1,
          1,
          1,
          1,
          1
      ],
      "n_dims_per_step": -1
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• eval_strategy: epoch
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 16
• gradient_accumulation_steps: 16
• learning_rate: 2e-05
• num_train_epochs: 10
• lr_scheduler_type: cosine
• warmup_ratio: 0.1
• tf32: False
• load_best_model_at_end: True
• optim: adamw_torch_fused
• batch_sampler: no_duplicates

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: epoch
• prediction_loss_only: True
• per_device_train_batch_size: 32
• per_device_eval_batch_size: 16
• per_gpu_train_batch_size: None
• per_gpu_eval_batch_size: None
• gradient_accumulation_steps: 16
• eval_accumulation_steps: None
• learning_rate: 2e-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: 10
• max_steps: -1
• lr_scheduler_type: cosine
• 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: False
• 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: True
• 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_fused
• 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
• batch_sampler: no_duplicates
• multi_dataset_batch_sampler: proportional

Training Logs

Epoch	Step	Training Loss	dim_128_cosine_map@100	dim_256_cosine_map@100	dim_512_cosine_map@100	dim_64_cosine_map@100	dim_768_cosine_map@100
0.9412	1	-	0.7980	0.8251	0.8141	0.7124	0.8260
1.8824	2	-	0.8624	0.8619	0.8691	0.7637	0.8557
2.8235	3	-	0.8763	0.8792	0.8770	0.8588	0.8832
3.7647	4	-	0.9007	0.9014	0.9115	0.8820	0.9130
4.7059	5	-	0.9014	0.9146	0.9186	0.9053	0.9185
5.6471	6	-	0.9134	0.9146	0.9186	0.9205	0.9183
6.5882	7	-	0.9255	0.9146	0.9186	0.9089	0.9185
7.5294	8	-	0.9255	0.9147	0.9186	0.9089	0.9185
8.4706	9	-	0.9255	0.9147	0.9186	0.9089	0.9186
9.4118	10	2.0337	0.9255	0.9148	0.9186	0.9089	0.9186

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.10.12
• Sentence Transformers: 3.0.1
• Transformers: 4.41.2
• PyTorch: 2.1.2+cu121
• Accelerate: 0.32.1
• Datasets: 2.19.1
• 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",
}

MatryoshkaLoss

@misc{kusupati2024matryoshka,
    title={Matryoshka Representation Learning}, 
    author={Aditya Kusupati and Gantavya Bhatt and Aniket Rege and Matthew Wallingford and Aditya Sinha and Vivek Ramanujan and William Howard-Snyder and Kaifeng Chen and Sham Kakade and Prateek Jain and Ali Farhadi},
    year={2024},
    eprint={2205.13147},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

@misc{henderson2017efficient,
    title={Efficient Natural Language Response Suggestion for Smart Reply}, 
    author={Matthew Henderson and Rami Al-Rfou and Brian Strope and Yun-hsuan Sung and Laszlo Lukacs and Ruiqi Guo and Sanjiv Kumar and Balint Miklos and Ray Kurzweil},
    year={2017},
    eprint={1705.00652},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}