<!-- 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-gutenberg-fixed
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0040
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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7298 | 0.29 | 500 | 5.6360 |
| 5.3656 | 0.58 | 1000 | 5.2026 |
| 5.0212 | 0.87 | 1500 | 4.9523 |
| 4.7476 | 1.16 | 2000 | 4.7988 |
| 4.586 | 1.45 | 2500 | 4.6801 |
| 4.4835 | 1.74 | 3000 | 4.5786 |
| 4.3674 | 2.03 | 3500 | 4.4991 |
| 4.1624 | 2.32 | 4000 | 4.4532 |
| 4.137 | 2.61 | 4500 | 4.3960 |
| 4.106 | 2.91 | 5000 | 4.3422 |
| 3.9133 | 3.2 | 5500 | 4.3427 |
| 3.8519 | 3.49 | 6000 | 4.3083 |
| 3.8433 | 3.78 | 6500 | 4.2794 |
| 3.758 | 4.07 | 7000 | 4.2761 |
| 3.5652 | 4.36 | 7500 | 4.2719 |
| 3.5749 | 4.65 | 8000 | 4.2517 |
| 3.5632 | 4.94 | 8500 | 4.2355 |
| 3.3622 | 5.23 | 9000 | 4.2584 |
| 3.3265 | 5.52 | 9500 | 4.2559 |
| 3.3112 | 5.81 | 10000 | 4.2500 |
| 3.264 | 6.1 | 10500 | 4.2572 |
| 3.1673 | 6.39 | 11000 | 4.2606 |
| 3.1623 | 6.68 | 11500 | 4.2607 |
| 3.1614 | 6.97 | 12000 | 4.2607 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
