distilgpt2-finetuned-general-stories
This model is a fine-tuned version of distilgpt2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.0063
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.7415 | 1.0 | 2940 | 5.4942 |
| 5.1216 | 2.0 | 5880 | 4.9520 |
| 4.7464 | 3.0 | 8820 | 4.6264 |
| 4.5081 | 4.0 | 11760 | 4.4148 |
| 4.3457 | 5.0 | 14700 | 4.2677 |
| 4.2207 | 6.0 | 17640 | 4.1709 |
| 4.1182 | 7.0 | 20580 | 4.0961 |
| 4.0599 | 8.0 | 23520 | 4.0473 |
| 4.0235 | 9.0 | 26460 | 4.0176 |
| 3.9931 | 10.0 | 29400 | 4.0063 |
Framework versions
- Transformers 4.40.0
- Pytorch 1.13.1+cu117
- Datasets 2.19.0
- Tokenizers 0.19.1