DoDucAnh/multilingual-e5-large-tuned

SentenceTransformer based on intfloat/multilingual-e5-large

This is a sentence-transformers model finetuned from intfloat/multilingual-e5-large on the mcqa_rag dataset. 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: intfloat/multilingual-e5-large
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 1024 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • mcqa_rag

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': False}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 1024, '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 = [
    'A student is mixing a cup of hot chocolate with a spoon. How is the heat transferred between the hot chocolate and the part of the spoon that is in the hot chocolate?\nA. Conduction transfers energy from the spoon to the hot chocolate.\nB. Conduction transfers energy from the hot chocolate to the spoon.\nC. Convection transfers energy from the spoon to the hot chocolate.\nD. Convection transfers energy from the hot chocolate to the spoon.',
    '**Applications of Convection in Everyday Life**: In mixtures like hot chocolate, convection plays a role when the liquid is stirred, causing hot liquid to rise and cooler liquid to sink, helping to mix and distribute heat throughout the drink. However, convection primarily pertains to the movement of the liquid rather than direct heat transfer from one solid to another.',
    'Additionally, moral philosophy often emphasizes the concept of empathy and the impact of one’s actions on others. Dwellings on a partner’s mistakes, while it may cause strain in a relationship, is often viewed as part of the natural dynamics of personal relationships, unless it leads to emotional abuse or significant harm. This indicates that emotional responses to insignificant errors may not always be deemed morally wrong, depending on context and impact.',
]
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]

Training Details

Training Dataset

mcqa_rag

• Dataset: mcqa_rag at 025e93d
• Size: 607,908 training samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 22 tokens mean: 105.96 tokens max: 512 tokens	min: 12 tokens mean: 70.95 tokens max: 478 tokens

• Samples:

  anchor	positive
  Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field. A. 0 B. 1 C. 2 D. 3	The notation Z_3 refers to the finite field with three elements, often denoted as {0, 1, 2}. This field operates under modular arithmetic, specifically modulo 3. Elements in Z_3 can be added and multiplied according to the rules of modulo 3, where any number can wrap around upon reaching 3.
  Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field. A. 0 B. 1 C. 2 D. 3	A field is a set equipped with two operations, addition and multiplication, satisfying certain properties: associativity, commutativity, distributivity, the existence of additive and multiplicative identities, and the existence of additive inverses and multiplicative inverses (for all elements except the zero element). In order for Z_3[x]/(f(x)) to be a field, the polynomial f(x) must be irreducible over Z_3.
  Find all c in Z_3 such that Z_3[x]/(x^2 + c) is a field. A. 0 B. 1 C. 2 D. 3	The expression Z_3[x] indicates the set of all polynomials with coefficients in Z_3. A polynomial is said to be irreducible over Z_3 if it cannot be factored into the product of two non-constant polynomials with coefficients in Z_3. In the case of quadratic polynomials like x^2 + c, irreducibility depends on whether it has any roots in the field Z_3.

• Loss: MultipleNegativesRankingLoss with these parameters:

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim"
  }
  

Evaluation Dataset

mcqa_rag

• Dataset: mcqa_rag at 025e93d
• Size: 14,380 evaluation samples
• Columns: anchor and positive
• Approximate statistics based on the first 1000 samples:

  	anchor	positive
  type	string	string
  details	min: 22 tokens mean: 99.4 tokens max: 512 tokens	min: 9 tokens mean: 59.56 tokens max: 501 tokens

• Samples:

  anchor	positive
  ക്രൂരകോഷ്ഠം ഉള്ള ഒരാളിൽ കോപിച്ചിരിക്കുന്ന ദോഷം താഴെപ്പറയുന്നവയിൽ ഏതാണ്? A. കഫം B. പിത്തം C. വാതം D. രക്തം	ഓരോ ദോഷത്തിനും അതിന്റേതായ സ്വഭാവങ്ങളും ശരീരത്തിൽ അത് ഉണ്ടാക്കുന്ന ഫലങ്ങളും ഉണ്ട്.
  Melyik tényező nem befolyásolja a fagylalt keresleti függvényét? A. A fagylalt árának változása. B. Mindegyik tényező befolyásolja. C. A jégkrém árának változása. D. A fagylalttölcsér árának változása.	A keresleti függvény negatív meredekségű, ami azt jelenti, hogy az ár növekedésével a keresett mennyiség csökken (csökkenő kereslet törvénye).
  In contrast to _______, _______ aim to reward favourable behaviour by companies. The success of such campaigns have been heightened through the use of ___________, which allow campaigns to facilitate the company in achieving _________ . A. Boycotts, Buyalls, Blockchain technology, Increased Sales B. Buycotts, Boycotts, Digital technology, Decreased Sales C. Boycotts, Buycotts, Digital technology, Decreased Sales D. Buycotts, Boycotts, Blockchain technology, Charitable donations E. Boycotts, Buyalls, Blockchain technology, Charitable donations F. Boycotts, Buycotts, Digital technology, Increased Sales G. Buycotts, Boycotts, Digital technology, Increased Sales H. Boycotts, Buycotts, Physical technology, Increased Sales I. Buycotts, Buyalls, Blockchain technology, Charitable donations J. Boycotts, Buycotts, Blockchain technology, Decreased Sales	Consumer Activism: This term refers to the actions taken by consumers to promote social, political, or environmental causes. These actions can include boycotting certain companies or buycotting others, influencing market dynamics based on ethical considerations. The effectiveness of consumer activism can vary but has gained prominence in recent years with increased visibility through social media.

• 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: 10
• per_device_eval_batch_size: 10
• learning_rate: 3e-05
• num_train_epochs: 2
• warmup_steps: 1000
• fp16: True
• load_best_model_at_end: True

All Hyperparameters

Click to expand

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: steps
• prediction_loss_only: True
• per_device_train_batch_size: 10
• per_device_eval_batch_size: 10
• 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: 3e-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: 2
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• warmup_steps: 1000
• 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: 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
• 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: None
• 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
0.1000	6080	0.099	0.0513
0.2000	12160	0.0811	0.0616
0.3000	18240	0.0765	0.0683
0.4001	24320	0.0722	0.0775
0.5001	30400	0.0676	0.0759
0.6001	36480	0.0641	0.0413
0.7001	42560	0.0559	0.0437
0.8001	48640	0.052	0.0581
0.9001	54720	0.0508	0.0540
1.0001	60800	0.0433	0.0402
1.1002	66880	0.0325	0.0350
1.2002	72960	0.0301	0.0379
1.3002	79040	0.0301	0.0355
1.4002	85120	0.027	0.0424
1.5002	91200	0.0251	0.0385
1.6002	97280	0.0218	0.0313
1.7003	103360	0.0205	0.0258
1.8003	109440	0.0201	0.0260
1.9003	115520	0.0181	0.0227

Framework Versions

• Python: 3.12.8
• Sentence Transformers: 3.4.1
• Transformers: 4.48.2
• PyTorch: 2.5.1+cu124
• Accelerate: 1.3.0
• Datasets: 3.2.0
• Tokenizers: 0.21.0

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}
}