legalclassBERT16

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

• Loss: 1.0753

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
1.9089	0.8	500	1.7507
1.6473	1.6	1000	1.5641
1.5486	2.4	1500	1.4464
1.478	3.2	2000	1.3976
1.3893	4.0	2500	1.3375
1.3631	4.8	3000	1.3016
1.296	5.6	3500	1.2723
1.2551	6.4	4000	1.2517
1.2417	7.2	4500	1.1834
1.2099	8.0	5000	1.1796
1.1763	8.8	5500	1.1516
1.1447	9.6	6000	1.1424
1.1298	10.4	6500	1.1258
1.0924	11.2	7000	1.1051
1.1158	12.0	7500	1.1098
1.097	12.8	8000	1.0857
1.0623	13.6	8500	1.0698
1.0505	14.4	9000	1.0803
1.0328	15.2	9500	1.0798
1.0263	16.0	10000	1.0753

Framework versions

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