Modelpublic

autumn10/sec-embedding-smoke

sourceHugging Faceupdated 5d agoView on Hugging Face

0likes34downloads

Model Card

SentenceTransformer based on unsloth/bge-m3

This is a sentence-transformers model finetuned from unsloth/bge-m3. It maps inputs to a 1024-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: unsloth/bge-m3
—Maximum Sequence Length: 1024 tokens
—Output Dimensionality: 1024 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': 'PeftModelForFeatureExtraction'})
  (1): Pooling({'embedding_dimension': 1024, 'pooling_mode': 'cls', 'include_prompt': True})
  (2): Normalize({'module_input_name': 'sentence_embedding', 'module_output_name': 'sentence_embedding'})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

bash

pip install -U sentence-transformers

Then you can load this model and run inference.

python

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("sentence_transformers_model_id")
# Run inference
queries = [
    'Các sản phẩm bị ảnh hưởng bởi CVE-2024-37242 có thể bị tấn công bằng cách nào?',
]
documents = [
    'CVE ID: CVE-2024-37242 | Cross-Site Request Forgery (CSRF) vulnerability in Automattic Newspack Newsletters newspack-newsletters allows Cross Site Request Forgery.This issue affects Newspack Newsletters: from n/a through <= 2.13.2. | Published: 2025-01-02',
    'CVE ID: CVE-2016-1000213 | Ruckus Wireless H500 web management interface CSRF | Published: 2016-10-25 | CVSS v3: 8.8 HIGH | Vector: CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H | AV:NETWORK AC:LOW PR:NONE UI:REQUIRED S:UNCHANGED | Impact: C:HIGH I:HIGH A:HIGH | CVSS v2: 6.8 | AV:N/AC:M/Au:N/C:P/I:P/A:P',
    'CVE ID: CVE-2003-1477 | MAILsweeper for SMTP 4.3.6 and 4.3.7 allows remote attackers to cause a denial of service (CPU consumption) via a PowerPoint attachment that either (1) is corrupt or (2) contains "embedded objects." | Published: 2003-12-31 | CVSS v2: 7.8 | AV:N/AC:L/Au:N/C:N/I:N/A:C',
]
query_embeddings = model.encode_query(queries)
document_embeddings = model.encode_document(documents)
print(query_embeddings.shape, document_embeddings.shape)
# [1, 1024] [3, 1024]

# Get the similarity scores for the embeddings
similarities = model.similarity(query_embeddings, document_embeddings)
print(similarities)
# tensor([[0.5969, 0.5004, 0.5534]])

<!--

Direct Usage (Transformers)

<details><summary>Click to see the direct usage in Transformers</summary>

</details> -->

<!--

Downstream Usage (Sentence Transformers)

You can finetune this model on your own dataset.

<details><summary>Click to expand</summary>

</details> -->

<!--

Out-of-Scope Use

List how the model may foreseeably be misused and address what users ought not to do with the model. -->

Evaluation

Metrics

Information Retrieval

—Dataset: validation
—Evaluated with <code>InformationRetrievalEvaluator</code>

Metric	Value
cosine_accuracy@1	0.8
cosine_accuracy@5	0.95
cosine_accuracy@10	1.0
cosine_precision@1	0.8
cosine_precision@5	0.19
cosine_precision@10	0.1
cosine_recall@1	0.8
cosine_recall@5	0.95
cosine_recall@10	1.0
cosine_ndcg@10	0.8803
cosine_mrr@10	0.8442
cosine_map@100	0.8442

<!--

Bias, Risks and Limitations

What are the known or foreseeable issues stemming from this model? You could also flag here known failure cases or weaknesses of the model. -->

<!--

Recommendations

What are recommendations with respect to the foreseeable issues? For example, filtering explicit content. -->

Training Details

Training Dataset

Unnamed Dataset

—Size: 50 training samples
—Columns: <code>anchor</code> and <code>positive</code>
—Approximate statistics based on the first 50 samples: | | anchor | positive | |:---------|:-----------------------------------------------------------------------------------|:-------------------------------------------------------------------------------------| | type | string | string | | modality | text | text | | details | <ul><li>min: 15 tokens</li><li>mean: 25.76 tokens</li><li>max: 39 tokens</li></ul> | <ul><li>min: 38 tokens</li><li>mean: 168.26 tokens</li><li>max: 364 tokens</li></ul> |
—Samples: | anchor | positive | |:----------------------------------------------------------------------------------------------------------------|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | <code>Các lỗ hổng tương tự CVE-2010-0629 trong các sản phẩm opensuse khác đã được công bố chưa?</code> | <code>CVE ID: CVE-2010-0629 \| Use-after-free vulnerability in kadmin/server/serverstubs.c in kadmind in MIT Kerberos 5 (aka krb5) 1.5 through 1.6.3 allows remote authenticated users to cause a denial of service (daemon crash) via a request from a kadmin client that sends an invalid API version number. \| Published: 2010-04-07 \| CVSS v3: 6.5 MEDIUM \| Vector: CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H \| AV:NETWORK AC:LOW PR:LOW UI:NONE S:UNCHANGED \| Impact: C:NONE I:NONE A:HIGH \| CVSS v2: 4.0 \| AV:N/AC:L/Au:S/C:N/I:N/A:P</code> | | <code>Cách kiểm tra xem hệ thống có bị ảnh hưởng bởi CVE-2003-1477 không, dựa trên sản phẩm allwindows?</code> | <code>CVE ID: CVE-2003-1477 \| MAILsweeper for SMTP 4.3.6 and 4.3.7 allows remote attackers to cause a denial of service (CPU consumption) via a PowerPoint attachment that either (1) is corrupt or (2) contains "embedded objects." \| Published: 2003-12-31 \| CVSS v2: 7.8 \| AV:N/AC:L/Au:N/C:N/I:N/A:C</code> | | <code>Có thông tin về thời gian phát hiện và công bố CVE-2005-3254 vào năm 2005 không?</code> | <code>CVE ID: CVE-2005-3254 \| The CGIwrap program before 3.9 on Debian GNU/Linux uses an incorrect minimum value of 100 for a UID to determine whether it can perform a seteuid operation, which could allow attackers to execute code as other system UIDs that are greater than the minimum value, which should be 1000 on Debian systems. \| Published: 2005-10-18 \| CVSS v2: 10.0 \| AV:N/AC:L/Au:N/C:C/I:C/A:C</code> |
—Loss: <code>CachedMultipleNegativesRankingLoss</code> with these parameters:

json

  {
      "scale": 20.0,
      "similarity_fct": "cos_sim",
      "mini_batch_size": 8,
      "mini_batch_num_tokens": null,
      "gather_across_devices": false,
      "directions": [
          "query_to_doc"
      ],
      "partition_mode": "joint",
      "hardness_mode": null,
      "hardness_strength": 0.0
  }

Training Hyperparameters

Non-Default Hyperparameters

—num_train_epochs: 1.0
—learning_rate: 2e-05
—bf16: True

All Hyperparameters

<details><summary>Click to expand</summary>

—per_device_train_batch_size: 8
—num_train_epochs: 1.0
—max_steps: -1
—learning_rate: 2e-05
—lr_scheduler_type: linear
—lr_scheduler_kwargs: None
—warmup_steps: 0
—optim: adamwtorchfused
—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: 1.0
—label_smoothing_factor: 0.0
—bf16: True
—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: trackio
—per_device_eval_batch_size: 8
—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: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
—parallelism_config: None
—dataloader_drop_last: False
—dataloader_num_workers: 0
—dataloader_pin_memory: True
—dataloader_persistent_workers: False
—dataloader_prefetch_factor: None
—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_backend: None
—ddp_timeout: 1800
—fsdp: []
—fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
—deepspeed: None
—debug: []
—skip_memory_metrics: True
—do_predict: False
—resume_from_checkpoint: None
—warmup_ratio: None
—local_rank: -1
—prompts: None
—batch_sampler: batch_sampler
—multi_dataset_batch_sampler: proportional
—router_mapping: {}
—learning_rate_mapping: {}

</details>

Training Logs

Epoch	Step	validation_cosine_ndcg@10
1.0	7	0.8803

Training Time

—Training: 4.2 seconds
—Evaluation: 0.2 seconds
—Total: 4.3 seconds

Framework Versions

—Python: 3.12.3
—Sentence Transformers: 6.0.1
—Transformers: 5.5.0
—PyTorch: 2.12.1+cu130
—Accelerate: 1.15.0
—Datasets: 4.3.0
—Tokenizers: 0.22.2

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

bibtex

@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",
}

CachedMultipleNegativesRankingLoss

bibtex

@misc{gao2021scaling,
    title={Scaling Deep Contrastive Learning Batch Size under Memory Limited Setup},
    author={Luyu Gao and Yunyi Zhang and Jiawei Han and Jamie Callan},
    year={2021},
    eprint={2101.06983},
    archivePrefix={arXiv},
    primaryClass={cs.LG}
}

MultipleNegativesRankingLoss

bibtex

@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},
}

<!--

Glossary

Clearly define terms in order to be accessible across audiences. -->

<!--

Model Card Authors

Lists the people who create the model card, providing recognition and accountability for the detailed work that goes into its construction. -->

<!--

Model Card Contact

Provides a way for people who have updates to the Model Card, suggestions, or questions, to contact the Model Card authors. -->