Modelpublic

qnguyenle/bge-attackqa-retriever

sourceHugging Faceupdated 5mo agoView on Hugging Face

0likes10downloads

Model Card

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

This is a sentence-transformers model finetuned from BAAI/bge-base-en-v1.5. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for retrieval.

Model Details

Model Description

—Model Type: Sentence Transformer
—Base model: BAAI/bge-base-en-v1.5
—Maximum Sequence Length: 512 tokens
—Output Dimensionality: 768 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': 768, 'pooling_mode': 'cls', 'include_prompt': True})
  (2): Normalize({})
)

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("qnguyenle/bge-attackqa-retriever")
# Run inference
sentences = [
    "Represent this sentence for searching relevant passages: What attack techniques are used by software 'S0144: ChChes'?",
    "The attack techniques used by software 'S0144: ChChes' are: 'T1036.005: Match Legitimate Name or Location', 'T1057: Process Discovery', 'T1071.001: Web Protocols', 'T1082: System Information Discovery', 'T1083: File and Directory Discovery', 'T1105: Ingress Tool Transfer', 'T1132.001: Standard Encoding', 'T1547.001: Registry Run Keys / Startup Folder', 'T1553.002: Code Signing', 'T1555.003: Credentials from Web Browsers', 'T1562.001: Disable or Modify Tools', 'T1573.001: Symmetric Cryptography'",
    "How attack software 'S1119: LIGHTWIRE' uses attack technique 'T1573.001: Symmetric Cryptography':\nLIGHTWIRE can RC4 encrypt C2 commands.",
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities)
# tensor([[1.0000, 0.9300, 0.0201],
#         [0.9300, 1.0000, 0.0798],
#         [0.0201, 0.0798, 1.0000]])

<!--

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. -->

<!--

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: 22,801 training samples
—Columns: <code>anchor</code> and <code>positive</code>
—Approximate statistics based on the first 1000 samples: | | anchor | positive | |:--------|:-----------------------------------------------------------------------------------|:------------------------------------------------------------------------------------| | type | string | string | | details | <ul><li>min: 14 tokens</li><li>mean: 32.93 tokens</li><li>max: 54 tokens</li></ul> | <ul><li>min: 18 tokens</li><li>mean: 80.87 tokens</li><li>max: 512 tokens</li></ul> |
—Samples: | anchor | positive | |:------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | <code>Represent this sentence for searching relevant passages: How does attack software 'S0074: Sakula' use attack technique 'T1543.003: Windows Service'?</code> | <code>How attack software 'S0074: Sakula' uses attack technique 'T1543.003: Windows Service':<br>Some Sakula samples install themselves as services for persistence by calling WinExec with the net start argument.</code> | | <code>Represent this sentence for searching relevant passages: What attack techniques are used by software 'S0386: Ursnif'?</code> | <code>The attack techniques used by software 'S0386: Ursnif' are: 'T1005: Data from Local System', 'T1007: System Service Discovery', 'T1012: Query Registry', 'T1027.010: Command Obfuscation', 'T1027.013: Encrypted/Encoded File', 'T1036.005: Match Legitimate Name or Location', 'T1041: Exfiltration Over C2 Channel', 'T1047: Windows Management Instrumentation', 'T1055.005: Thread Local Storage', 'T1055.012: Process Hollowing', 'T1056.004: Credential API Hooking', 'T1057: Process Discovery', 'T1059.001: PowerShell', 'T1059.005: Visual Basic', 'T1070.004: File Deletion', 'T1071.001: Web Protocols', 'T1074.001: Local Data Staging', 'T1080: Taint Shared Content', 'T1082: System Information Discovery', 'T1090.003: Multi-hop Proxy', 'T1090: Proxy', 'T1091: Replication Through Removable Media', 'T1105: Ingress Tool Transfer', 'T1106: Native API', 'T1112: Modify Registry', 'T1113: Screen Capture', 'T1132: Data Encoding', 'T1140: Deobfuscate/Decode Files or Information', 'T1185: Browser Session Hijacki...</code> | | <code>Represent this sentence for searching relevant passages: How does attack software 'S0435: PLEAD' use attack technique 'T1555: Credentials from Password Stores'?</code> | <code>How attack software 'S0435: PLEAD' uses attack technique 'T1555: Credentials from Password Stores':<br>PLEAD has the ability to steal saved passwords from Microsoft Outlook.</code> |
—Loss: <code>MultipleNegativesRankingLoss</code> with these parameters:

json

  {
      "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
—learning_rate: 2e-05
—num_train_epochs: 1
—warmup_steps: 0.1
—fp16: True

All Hyperparameters

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

—do_predict: False
—eval_strategy: no
—prediction_loss_only: True
—per_device_train_batch_size: 32
—per_device_eval_batch_size: 8
—gradient_accumulation_steps: 1
—eval_accumulation_steps: None
—torch_empty_cache_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: 1
—max_steps: -1
—lr_scheduler_type: linear
—lr_scheduler_kwargs: None
—warmup_ratio: None
—warmup_steps: 0.1
—log_level: passive
—log_level_replica: warning
—log_on_each_node: True
—logging_nan_inf_filter: True
—enable_jit_checkpoint: False
—save_on_each_node: False
—save_only_model: False
—restore_callback_states_from_checkpoint: False
—use_cpu: False
—seed: 42
—data_seed: None
—bf16: False
—fp16: True
—bf16_full_eval: False
—fp16_full_eval: False
—tf32: None
—local_rank: -1
—ddp_backend: None
—debug: []
—dataloader_drop_last: False
—dataloader_num_workers: 0
—dataloader_prefetch_factor: None
—disable_tqdm: False
—remove_unused_columns: True
—label_names: None
—load_best_model_at_end: False
—ignore_data_skip: False
—fsdp: []
—fsdp_config: {'minnumparams': 0, 'xla': False, 'xlafsdpv2': False, 'xlafsdpgrad_ckpt': False}
—accelerator_config: {'splitbatches': False, 'dispatchbatches': None, 'evenbatches': True, 'useseedablesampler': True, 'nonblocking': False, 'gradientaccumulationkwargs': None}
—parallelism_config: None
—deepspeed: None
—label_smoothing_factor: 0.0
—optim: adamwtorchfused
—optim_args: None
—group_by_length: False
—length_column_name: length
—project: huggingface
—trackio_space_id: trackio
—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
—push_to_hub: False
—resume_from_checkpoint: None
—hub_model_id: None
—hub_strategy: every_save
—hub_private_repo: None
—hub_always_push: False
—hub_revision: None
—gradient_checkpointing: False
—gradient_checkpointing_kwargs: None
—include_for_metrics: []
—eval_do_concat_batches: True
—auto_find_batch_size: False
—full_determinism: False
—ddp_timeout: 1800
—torch_compile: False
—torch_compile_backend: None
—torch_compile_mode: None
—include_num_input_tokens_seen: no
—neftune_noise_alpha: None
—optim_target_modules: None
—batch_eval_metrics: False
—eval_on_start: False
—use_liger_kernel: False
—liger_kernel_config: None
—eval_use_gather_object: False
—average_tokens_across_devices: True
—use_cache: False
—prompts: None
—batch_sampler: batch_sampler
—multi_dataset_batch_sampler: proportional
—router_mapping: {}
—learning_rate_mapping: {}

</details>

Training Logs

Epoch	Step	Training Loss
0.0701	50	0.1743
0.1403	100	0.0247
0.2104	150	0.0204
0.2805	200	0.0178
0.3506	250	0.0134
0.4208	300	0.0122
0.4909	350	0.0127
0.5610	400	0.0161
0.6311	450	0.0127
0.7013	500	0.0096
0.7714	550	0.0127
0.8415	600	0.0119
0.9116	650	0.0140
0.9818	700	0.0125

Training Time

—Training: 11.3 minutes

Framework Versions

—Python: 3.12.13
—Sentence Transformers: 5.4.0
—Transformers: 5.0.0
—PyTorch: 2.10.0+cu128
—Accelerate: 1.13.0
—Datasets: 4.0.0
—Tokenizers: 0.22.2

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",
}

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. -->