BERT-Indian-legal-NER

This model is a fine-tuned version of google-bert/bert-base-cased on hf-tuner/indian-legal-ner dataset. It achieves the following results on the evaluation set:

• Loss: 0.1296

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 2e-05
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
0.1857	0.6431	1000	0.1691
0.1181	1.2862	2000	0.1298
0.1372	1.9293	3000	0.1279
0.0688	2.5723	4000	0.1296

Framework versions

• Transformers 4.57.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1