<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
gpt2-concat-aochildes-length-15k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1875
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: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7208 | 0.29 | 500 | 5.6413 |
| 5.3798 | 0.59 | 1000 | 5.2022 |
| 5.026 | 0.88 | 1500 | 4.9544 |
| 4.7535 | 1.18 | 2000 | 4.8031 |
| 4.5938 | 1.47 | 2500 | 4.6839 |
| 4.4847 | 1.76 | 3000 | 4.5811 |
| 4.3568 | 2.06 | 3500 | 4.5046 |
| 4.1613 | 2.35 | 4000 | 4.4593 |
| 4.1394 | 2.65 | 4500 | 4.4021 |
| 4.0897 | 2.94 | 5000 | 4.3497 |
| 3.874 | 3.24 | 5500 | 4.3454 |
| 3.8331 | 3.53 | 6000 | 4.3191 |
| 3.8104 | 3.82 | 6500 | 4.2890 |
| 3.6885 | 4.12 | 7000 | 4.2909 |
| 3.5369 | 4.41 | 7500 | 4.2866 |
| 3.5339 | 4.71 | 8000 | 4.2735 |
| 3.5159 | 5.0 | 8500 | 4.2598 |
| 3.3458 | 5.29 | 9000 | 4.2780 |
| 3.3397 | 5.59 | 9500 | 4.2764 |
| 3.3365 | 5.88 | 10000 | 4.2765 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
