SentenceTransformer based on BAAI/bge-large-en-v1.5

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 dimensions
  • Similarity Function: Cosine Similarity

Model Sources

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("chinchilla04/bge-finetuned-train")
# Run inference
sentences = [
    'how do i ask about the weather in chinese',
    'Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation.',
    "Requests for information about a vehicle's miles per gallon (MPG) rating, either in specific conditions like city driving or as an overall performance metric. Users may seek guidance on fuel efficiency for their car.",
]
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

Metric Value
cosine_accuracy@1 0.9707
cosine_accuracy@3 0.9887
cosine_accuracy@5 0.992
cosine_accuracy@10 0.9957
cosine_precision@1 0.9707
cosine_precision@3 0.3296
cosine_precision@5 0.1984
cosine_precision@10 0.0996
cosine_recall@1 0.9707
cosine_recall@3 0.9887
cosine_recall@5 0.992
cosine_recall@10 0.9957
cosine_ndcg@10 0.9842
cosine_mrr@10 0.9804
cosine_map@100 0.9806

Training Details

Training Dataset

Unnamed Dataset

  • Size: 15,002 training samples
  • Columns: anchor, positive, and negative
  • Approximate statistics based on the first 1000 samples:
    anchor positive negative
    type string string string
    details
    • min: 4 tokens
    • mean: 10.66 tokens
    • max: 28 tokens
    • min: 25 tokens
    • mean: 42.6 tokens
    • max: 58 tokens
    • min: 29 tokens
    • mean: 41.95 tokens
    • max: 58 tokens
  • Samples:
    anchor positive negative
    what expression would i use to say i love you if i were an italian Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation. Requests involving financial operations, such as transferring money between bank accounts, credit cards, or other financial instruments. Users typically specify the amount, the source account, and the target account, ensuring that the transfer is executed correctly and securely.
    can you tell me how to say 'i do not speak much spanish', in spanish Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation. Requests involving financial operations, such as transferring money between bank accounts, credit cards, or other financial instruments. Users typically specify the amount, the source account, and the target account, ensuring that the transfer is executed correctly and securely.
    what is the equivalent of, 'life is good' in french Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation. Requests involving financial operations, such as transferring money between bank accounts, credit cards, or other financial instruments. Users typically specify the amount, the source account, and the target account, ensuring that the transfer is executed correctly and securely.
  • Loss: MultipleNegativesRankingLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "cos_sim"
    }
    

Evaluation Dataset

Unnamed Dataset

  • Size: 3,000 evaluation samples
  • Columns: anchor and positive
  • Approximate statistics based on the first 1000 samples:
    anchor positive
    type string string
    details
    • min: 3 tokens
    • mean: 11.06 tokens
    • max: 29 tokens
    • min: 26 tokens
    • mean: 36.16 tokens
    • max: 58 tokens
  • Samples:
    anchor positive
    in spanish, meet me tomorrow is said how Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation.
    in french, how do i say, see you later Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation.
    how do you say hello in japanese Requests related to translating words, phrases, or sentences from one language to another. The user may specify the source and target languages, and the goal is to provide an accurate and context-appropriate translation.
  • Loss: MultipleNegativesRankingLoss with these parameters:
    {
        "scale": 20.0,
        "similarity_fct": "cos_sim"
    }
    

Training Hyperparameters

Non-Default Hyperparameters

  • eval_strategy: steps
  • per_device_train_batch_size: 32
  • learning_rate: 1e-05
  • num_train_epochs: 4
  • lr_scheduler_type: cosine
  • warmup_ratio: 0.2
  • load_best_model_at_end: True
  • optim: adamw_torch_fused

All Hyperparameters

Click to expand
  • overwrite_output_dir: False
  • do_predict: False
  • eval_strategy: steps
  • prediction_loss_only: True
  • per_device_train_batch_size: 32
  • per_device_eval_batch_size: 8
  • 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: 1e-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: cosine
  • lr_scheduler_kwargs: {}
  • warmup_ratio: 0.2
  • 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: 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
  • 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 cosine_ndcg@10
None 0 - 0.2730 0.9055
0.3198 150 - 0.0698 0.9633
0.6397 300 - 0.0642 0.9683
0.9595 450 - 0.0603 0.9763
1.0661 500 1.0338 - -
1.2793 600 - 0.0612 0.9762
1.5991 750 - 0.0602 0.9802
1.9190 900 - 0.0571 0.9820
2.1322 1000 0.787 - -
2.2388 1050 - 0.0585 0.9819
2.5586 1200 - 0.0565 0.9842
2.8785 1350 - 0.0578 0.9837
3.1983 1500 0.6768 0.0570 0.9844
3.5181 1650 - 0.0587 0.9837
3.8380 1800 - 0.0584 0.9837
None 0 - 0.0565 0.9842
  • The bold row denotes the saved checkpoint.

Framework Versions

  • Python: 3.10.14
  • Sentence Transformers: 3.3.1
  • Transformers: 4.46.3
  • PyTorch: 2.4.0
  • Accelerate: 1.1.1
  • Datasets: 3.1.0
  • Tokenizers: 0.20.3

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{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}
}
Downloads last month
228
Safetensors
Model size
335M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for chinchilla04/bge-finetuned-train

Finetuned
(24)
this model

Evaluation results