cl-length-260k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.8321
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: 64
- eval_batch_size: 64
- seed: 42
- 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 |
|---|---|---|---|
| 5.8312 | 0.06 | 500 | 5.5636 |
| 4.6153 | 0.11 | 1000 | 5.2363 |
| 4.354 | 0.17 | 1500 | 5.0666 |
| 4.174 | 0.23 | 2000 | 4.9628 |
| 4.05 | 0.28 | 2500 | 4.9068 |
| 3.9481 | 0.34 | 3000 | 4.8695 |
| 3.8506 | 0.39 | 3500 | 4.8265 |
| 3.7636 | 0.45 | 4000 | 4.8014 |
| 3.6826 | 0.51 | 4500 | 4.7895 |
| 3.5968 | 0.56 | 5000 | 4.7650 |
| 3.5137 | 0.62 | 5500 | 4.7536 |
| 3.4435 | 0.68 | 6000 | 4.7414 |
| 3.3713 | 0.73 | 6500 | 4.7371 |
| 3.3116 | 0.79 | 7000 | 4.7364 |
| 3.2597 | 0.84 | 7500 | 4.7324 |
| 3.2291 | 0.9 | 8000 | 4.7300 |
| 3.2165 | 0.96 | 8500 | 4.7260 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
- Downloads last month
- 3