BERT-legal-de-cased_German_legal_SQuAD_17

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 4.2733

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-05
• train_batch_size: 160
• eval_batch_size: 40
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 17

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	2	6.1900
No log	2.0	4	6.1896
No log	3.0	6	6.2832
No log	4.0	8	6.1739
No log	5.0	10	5.8089
No log	6.0	12	5.5485
No log	7.0	14	5.3540
No log	8.0	16	5.1463
No log	9.0	18	4.9179
No log	10.0	20	4.7521
No log	11.0	22	4.6237
No log	12.0	24	4.5150
No log	13.0	26	4.4347
No log	14.0	28	4.3646
No log	15.0	30	4.3187
No log	16.0	32	4.2865
No log	17.0	34	4.2733

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu121
• Datasets 2.14.7
• Tokenizers 0.15.0