GQA_RoBERTa_German_legal_SQuAD_17

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

• Loss: 1.3681

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: 128
• eval_batch_size: 32
• 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	3	4.0170
No log	2.0	6	3.5344
No log	3.0	9	2.9885
No log	4.0	12	2.6694
No log	5.0	15	2.4544
No log	6.0	18	2.2128
No log	7.0	21	2.0608
No log	8.0	24	1.9519
No log	9.0	27	1.7820
No log	10.0	30	1.6573
No log	11.0	33	1.6004
No log	12.0	36	1.5343
No log	13.0	39	1.4756
No log	14.0	42	1.4266
No log	15.0	45	1.3978
No log	16.0	48	1.3776
No log	17.0	51	1.3681

Framework versions

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