GQA_RoBERTa
This model was trained from scratch on the germanquad dataset.
It achieves the following results on the evaluation set:
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|
| No log | 1.0 | 90 | 5.4929 |
| No log | 2.0 | 180 | 5.5185 |
| No log | 3.0 | 270 | 5.3102 |
| No log | 4.0 | 360 | 4.9559 |
| No log | 5.0 | 450 | 4.6422 |
| 4.3295 | 6.0 | 540 | 4.3858 |
| 4.3295 | 7.0 | 630 | 4.1523 |
| 4.3295 | 8.0 | 720 | 4.0114 |
| 4.3295 | 9.0 | 810 | 3.5878 |
| 4.3295 | 10.0 | 900 | 3.6762 |
| 4.3295 | 11.0 | 990 | 3.7009 |
| 2.719 | 12.0 | 1080 | 3.7030 |
| 2.719 | 13.0 | 1170 | 3.9876 |
| 2.719 | 14.0 | 1260 | 3.7367 |
| 2.719 | 15.0 | 1350 | 3.7522 |
| 2.719 | 16.0 | 1440 | 3.6393 |
| 2.0323 | 17.0 | 1530 | 3.7502 |
| 2.0323 | 18.0 | 1620 | 3.8672 |
| 2.0323 | 19.0 | 1710 | 3.8356 |
| 2.0323 | 20.0 | 1800 | 3.8389 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.14.7
- Tokenizers 0.15.0
