gpt2-small-finetuned-codeparrot-ds_nlp-course-chapter7-section5
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0606
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: 0.0005
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5691 | 0.08 | 5000 | 1.7463 |
| 1.6818 | 0.15 | 10000 | 1.5248 |
| 1.5328 | 0.23 | 15000 | 1.4226 |
| 1.4521 | 0.31 | 20000 | 1.3569 |
| 1.3944 | 0.38 | 25000 | 1.3030 |
| 1.3422 | 0.46 | 30000 | 1.2558 |
| 1.2976 | 0.54 | 35000 | 1.2129 |
| 1.2514 | 0.61 | 40000 | 1.1714 |
| 1.2089 | 0.69 | 45000 | 1.1321 |
| 1.1737 | 0.77 | 50000 | 1.0990 |
| 1.1427 | 0.84 | 55000 | 1.0758 |
| 1.1242 | 0.92 | 60000 | 1.0636 |
| 1.1142 | 1.0 | 65000 | 1.0606 |
Framework versions
- Transformers 4.35.2
- Pytorch 1.11.0+cu102
- Datasets 2.15.0
- Tokenizers 0.15.0