fine-tuned-legal-bert

SentenceTransformer based on nlpaueb/legal-bert-base-uncased

This is a sentence-transformers model finetuned from nlpaueb/legal-bert-base-uncased. 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: nlpaueb/legal-bert-base-uncased <!-- at revision 15b570cbf88259610b082a167dacc190124f60f6 -->
• Maximum Sequence Length: 512 tokens
• Output Dimensionality: 768 dimensions
• Similarity Function: Cosine Similarity

<!-- - **Training Dataset:** Unknown --> <!-- - **Language:** Unknown --> <!-- - **License:** Unknown -->

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, 'architecture': 'BertModel'})
  (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})
)

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("Tushar0505/fine-tuned-legal-bert")
# Run inference
sentences = [
    "CC\n565 [LNIND 1994 SC 952] : 1995 SCC (Cr) 29, quoted from D F Marion v Davis, 1989 (1) SCC 494\n[LNIND 1989 SC 10] : AIR 1989 SC 714 [LNIND 1989 SC 833] .\n4. Kishore Samrite v State of UP, (2013) 2 SCC 398 [LNIND 2012 SC 657] .\n5. Subramanian Swamy v UOI, Ministry of Law, 2016 Cr LJ 3214 : 2016 (5) SCJ 643 .'reputation' is what he is supposed to be in what people say he is. 'Character' depends\non attributes possessed and 'reputation' on attributes which others believe one to\npossess. The former ",
    "CC\n565 [LNIND 1994 SC 952] : 1995 SCC (Cr) 29, quoted from D F Marion v Davis, 1989 (1) SCC 494\n[LNIND 1989 SC 10] : AIR 1989 SC 714 [LNIND 1989 SC 833] .\n4. Kishore Samrite v State of UP, (2013) 2 SCC 398 [LNIND 2012 SC 657] .\n5. Subramanian Swamy v UOI, Ministry of Law, 2016 Cr LJ 3214 : 2016 (5) SCJ 643 .'reputation' is what he is supposed to be in what people say he is. 'Character' depends\non attributes possessed and 'reputation' on attributes which others believe one to\npossess. The former ",
    'of any other reliable evidence.40. In a case where the agreement is for accomplishment\nof an act which by itself constitutes an offence, then in that event, unless the Statute so\nrequires, no overt act is necessary to be proved by the prosecution because in such a\nfact-situation criminal conspiracy is established by proving such an agreement.41.\nWhen two agree to carry it into effect, the very plot is an act in itself, and an act of each\nof the parties, promise against promise, actus contra actu',
]
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, 1.0000, 0.9928],
#         [1.0000, 1.0000, 0.9928],
#         [0.9928, 0.9928, 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: 10,048 training samples
• Columns: <code>sentence_0</code>, <code>sentence_1</code>, and <code>label</code>
• Approximate statistics based on the first 1000 samples:

  	sentence_0	sentence_1	label
  type	string	string	float
  details	<ul><li>min: 89 tokens</li><li>mean: 124.6 tokens</li><li>max: 241 tokens</li></ul>	<ul><li>min: 89 tokens</li><li>mean: 124.6 tokens</li><li>max: 241 tokens</li></ul>	<ul><li>min: 1.0</li><li>mean: 1.0</li><li>max: 1.0</li></ul>

• Samples:

  sentence_0	sentence_1	label
  <code>xecution of<br>the sentence eventually confirmed. This custody is different from custody of a convict<br>suffering from simple or rigorous imprisonment. Therefore, he cannot be kept in Cell or<br>solitary confinement which itself is a separate punishment.246.<br>[s 73.1] Convicts on death row.—<br>The question about solitary confinement or keeping condemned prisoner alone under<br>strict guard as provided in various jail manual has been considered in depth by the<br>Constitution Bench in Sunil Batra v Delhi Administrat</code>	<code>xecution of<br>the sentence eventually confirmed. This custody is different from custody of a convict<br>suffering from simple or rigorous imprisonment. Therefore, he cannot be kept in Cell or<br>solitary confinement which itself is a separate punishment.246.<br>[s 73.1] Convicts on death row.—<br>The question about solitary confinement or keeping condemned prisoner alone under<br>strict guard as provided in various jail manual has been considered in depth by the<br>Constitution Bench in Sunil Batra v Delhi Administrat</code>	<code>1.0</code>
  <code>Only a house-trespass committed under such circumstances as may reasonably cause<br>apprehension that death or grievous hurt would be the consequence is enumerated as<br>one of the offences under section 103.378. But in another judgment Supreme Court<br>held that there is no law that right of self-defence cannot be exercised in relation to a<br>dispute over an open space.379. Where the trespasser was in settled possession of the<br>field in question, and the party who claimed ownership started ploughing the fiel</code>	<code>Only a house-trespass committed under such circumstances as may reasonably cause<br>apprehension that death or grievous hurt would be the consequence is enumerated as<br>one of the offences under section 103.378. But in another judgment Supreme Court<br>held that there is no law that right of self-defence cannot be exercised in relation to a<br>dispute over an open space.379. Where the trespasser was in settled possession of the<br>field in question, and the party who claimed ownership started ploughing the fiel</code>	<code>1.0</code>
  <code>ns (sections 231, 232).<br>(2)Making or selling instrument for counterfeiting (sections 233, 234).<br>(3)Possession of instrument for counterfeiting (section 235).<br>(4)Abetting in India the counterfeiting of coin out of India (section 236).<br>(5)Importing or exporting of counterfeit coin (sections 237, 238).<br>(6)Delivering counterfeit coin knowing it to be so (sections 239, 240).<br>(7)Delivering counterfeit coin not known to be so when first possessed<br>(section 241).<br>(8)Possession of counterfeit coin knowing </code>	<code>ns (sections 231, 232).<br>(2)Making or selling instrument for counterfeiting (sections 233, 234).<br>(3)Possession of instrument for counterfeiting (section 235).<br>(4)Abetting in India the counterfeiting of coin out of India (section 236).<br>(5)Importing or exporting of counterfeit coin (sections 237, 238).<br>(6)Delivering counterfeit coin knowing it to be so (sections 239, 240).<br>(7)Delivering counterfeit coin not known to be so when first possessed<br>(section 241).<br>(8)Possession of counterfeit coin knowing </code>	<code>1.0</code>

• Loss: <code>CosineSimilarityLoss</code> with these parameters:

  {
      "loss_fct": "torch.nn.modules.loss.MSELoss"
  }
  

Training Hyperparameters

Non-Default Hyperparameters

• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• num_train_epochs: 2
• multi_dataset_batch_sampler: round_robin

All Hyperparameters

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

• overwrite_output_dir: False
• do_predict: False
• eval_strategy: no
• prediction_loss_only: True
• per_device_train_batch_size: 16
• per_device_eval_batch_size: 16
• 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: 5e-05
• weight_decay: 0.0
• adam_beta1: 0.9
• adam_beta2: 0.999
• adam_epsilon: 1e-08
• max_grad_norm: 1
• num_train_epochs: 2
• max_steps: -1
• lr_scheduler_type: linear
• lr_scheduler_kwargs: {}
• warmup_ratio: 0.0
• 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: False
• 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}
• parallelism_config: 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: None
• hub_always_push: False
• hub_revision: None
• 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
• 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
• liger_kernel_config: None
• eval_use_gather_object: False
• average_tokens_across_devices: False
• prompts: None
• batch_sampler: batch_sampler
• multi_dataset_batch_sampler: round_robin
• router_mapping: {}
• learning_rate_mapping: {}

</details>

Training Logs

Framework Versions

• Python: 3.12.11
• Sentence Transformers: 5.1.0
• Transformers: 4.56.1
• PyTorch: 2.8.0+cu126
• Accelerate: 1.10.1
• Datasets: 4.0.0
• Tokenizers: 0.22.0

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",
}

<!-- ## 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.* -->