bert-leg-al-corpus

This model is a fine-tuned version of joelniklaus/legal-xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.7948

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: 2e-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
1.8679	0.4219	100	1.8208
1.8623	0.8439	200	1.8081
1.8987	1.2658	300	1.8153
1.8659	1.6878	400	1.7836
1.8685	2.1097	500	1.7825
1.8734	2.5316	600	1.7903
1.9124	2.9536	700	1.7771
1.8875	3.3755	800	1.7819
1.9029	3.7975	900	1.8032

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1