modernbert-embed-base-legal-original

ModernBERT Embed base Legal Matryoshka

This is a sentence-transformers model finetuned from nomic-ai/modernbert-embed-base on the json dataset. It maps sentences & paragraphs to a 768-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: nomic-ai/modernbert-embed-base <!-- at revision d556a88e332558790b210f7bdbe87da2fa94a8d8 -->
• Maximum Sequence Length: 8192 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity
• Training Dataset:
  • json
• 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': 8192, 'do_lower_case': False, 'architecture': 'ModernBertModel'})
  (1): Pooling({'word_embedding_dimension': 768, '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("aaa961/modernbert-embed-base-legal-original")
# Run inference
sentences = [
    'conclusion, however, is weighty—steeped in myriad complexity and fraught with tension—and in the Court’s view, \nthis conclusion has significant implications for the scope of the FOIA.  The Court will further discuss the two-fold \nreasoning that leads to this result. \nFirst, permitting a member of the public to request from an agency a listing of search results or a listing that',
    'What does the Court believe about the conclusion?',
    'Where can the statement about the best value basis for awards in Polaris be found?',
]
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, 0.5555, 0.0863],
#         [0.5555, 1.0000, 0.1753],
#         [0.0863, 0.1753, 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.* -->

Evaluation

Metrics

Information Retrieval

• Dataset: dim_768
• Evaluated with <code>InformationRetrievalEvaluator</code> with these parameters:

  {
      "truncate_dim": 768
  }
  

Information Retrieval

• Dataset: dim_512
• Evaluated with <code>InformationRetrievalEvaluator</code> with these parameters:

  {
      "truncate_dim": 512
  }
  

Information Retrieval

• Dataset: dim_256
• Evaluated with <code>InformationRetrievalEvaluator</code> with these parameters:

  {
      "truncate_dim": 256
  }
  

Information Retrieval

• Dataset: dim_128
• Evaluated with <code>InformationRetrievalEvaluator</code> with these parameters:

  {
      "truncate_dim": 128
  }
  

Information Retrieval

• Dataset: dim_64
• Evaluated with <code>InformationRetrievalEvaluator</code> with these parameters:

  {
      "truncate_dim": 64
  }
  

<!-- ## 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

json

• Dataset: json
• Size: 5,822 training samples
• Columns: <code>positive</code> and <code>anchor</code>
• Approximate statistics based on the first 1000 samples:

  	positive	anchor
  type	string	string
  details	<ul><li>min: 33 tokens</li><li>mean: 97.83 tokens</li><li>max: 160 tokens</li></ul>	<ul><li>min: 8 tokens</li><li>mean: 16.69 tokens</li><li>max: 38 tokens</li></ul>

• Samples:

  positive	anchor
  <code>the IRGs. Id. at 248 & n.15. It did not matter that the NIMH “may be greatly influenced” by an <br>IRG’s “expert view.” Id. at 248. Given the functions that IRGs were “empowered by law to <br>perform,” they did not wield “substantial independent authority.” Id. at 247–48. <br> <br>Two months after Washington Research Project, Congress enacted the 1974 amendment</code>	<code>What did Congress enact two months after Washington Research Project?</code>
  <code>GSA’s interpretation of 13 C.F.R. § 125.9(b)(3)(i) harms protégés has broad implications. If <br>exclusion from bidding on the SB Solicitation indeed harms either protégé member of SHS or <br>VCH, perhaps this suggests the mentor-protégé relationships should not have been approved in the <br>first instance. See 13 C.F.R. § 125.9(b)(3) (“In order for SBA to agree to allow a mentor to have</code>	<code>Which two protégés could be harmed by exclusion from bidding on the SB Solicitation?</code>
  <code>Black’s Law Dictionary 742 (9th ed. 2009) (defining “function” as “[a]ctivity that is appropriate <br>to a particular business or profession”); Webster’s Third New Int’l Dictionary 920 (1981) <br>(defining “function” as “the action for which a person or thing is specially fitted, used, or <br>responsible or for which a thing exists”).</code>	<code>What year was the 9th edition of Black’s Law Dictionary published?</code>

• Loss: <code>MatryoshkaLoss</code> 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

• per_device_train_batch_size: 32
• num_train_epochs: 4
• learning_rate: 2e-05
• lr_scheduler_type: cosine
• warmup_steps: 0.1
• optim: adamw_torch_fused
• gradient_accumulation_steps: 16
• bf16: True
• tf32: True
• eval_strategy: epoch
• per_device_eval_batch_size: 16
• load_best_model_at_end: True
• batch_sampler: no_duplicates

All Hyperparameters

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

• per_device_train_batch_size: 32
• num_train_epochs: 4
• max_steps: -1
• learning_rate: 2e-05
• lr_scheduler_type: cosine
• lr_scheduler_kwargs: None
• warmup_steps: 0.1
• optim: adamw_torch_fused
• 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: 16
• 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: True
• 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
• eval_strategy: epoch
• per_device_eval_batch_size: 16
• 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: True
• ignore_data_skip: False
• restore_callback_states_from_checkpoint: False
• full_determinism: False
• seed: 42
• data_seed: None
• use_cpu: False
• accelerator_config: {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': 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: {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_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: no_duplicates
• multi_dataset_batch_sampler: proportional
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

• The bold row denotes the saved checkpoint.

Framework Versions

• Python: 3.12.11
• Sentence Transformers: 5.3.0
• Transformers: 5.3.0
• PyTorch: 2.5.1+cu121
• Accelerate: 1.13.0
• Datasets: 4.8.2
• Tokenizers: 0.22.2

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{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.* -->