legalBERTclass16

This model is a fine-tuned version of nlpaueb/legal-bert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8713

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 16

Training results

Training Loss	Epoch	Step	Validation Loss
1.4448	0.8	500	1.3321
1.2908	1.6	1000	1.2269
1.2095	2.4	1500	1.1415
1.1719	3.2	2000	1.1189
1.1117	4.0	2500	1.0708
1.0734	4.8	3000	1.0739
1.0157	5.6	3500	1.0240
0.9984	6.4	4000	0.9927
0.9614	7.2	4500	0.9533
0.9567	8.0	5000	0.9564
0.9296	8.8	5500	0.9449
0.9035	9.6	6000	0.9122
0.8902	10.4	6500	0.9244
0.8459	11.2	7000	0.8759
0.8616	12.0	7500	0.9022
0.8317	12.8	8000	0.8729
0.8307	13.6	8500	0.8655
0.8296	14.4	9000	0.8614
0.8063	15.2	9500	0.8714
0.8185	16.0	10000	0.8713

Framework versions

• Transformers 4.49.0.dev0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0