distilgpt2-ft
This model is a fine-tuned version of distilgpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.3824
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: 0.000166
- 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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 16 | 2.2852 |
| No log | 2.0 | 32 | 2.2098 |
| No log | 3.0 | 48 | 2.2370 |
| No log | 4.0 | 64 | 2.3000 |
| No log | 5.0 | 80 | 2.3898 |
| No log | 6.0 | 96 | 2.4586 |
| No log | 7.0 | 112 | 2.5484 |
| No log | 8.0 | 128 | 2.6572 |
| No log | 9.0 | 144 | 2.7703 |
| No log | 10.0 | 160 | 2.9010 |
| No log | 11.0 | 176 | 2.9734 |
| No log | 12.0 | 192 | 3.0461 |
| No log | 13.0 | 208 | 3.1837 |
| No log | 14.0 | 224 | 3.2359 |
| No log | 15.0 | 240 | 3.2506 |
| No log | 16.0 | 256 | 3.2979 |
| No log | 17.0 | 272 | 3.3512 |
| No log | 18.0 | 288 | 3.3811 |
| No log | 19.0 | 304 | 3.3787 |
| No log | 20.0 | 320 | 3.3824 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3