<!-- 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-2p2k-1k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0101
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 |
|---|---|---|---|
| 6.7263 | 0.29 | 500 | 5.6343 |
| 5.3716 | 0.58 | 1000 | 5.2005 |
| 5.0162 | 0.88 | 1500 | 4.9564 |
| 4.7483 | 1.17 | 2000 | 4.8083 |
| 4.5898 | 1.46 | 2500 | 4.6842 |
| 4.484 | 1.75 | 3000 | 4.5777 |
| 4.3681 | 2.04 | 3500 | 4.4955 |
| 4.1667 | 2.33 | 4000 | 4.4513 |
| 4.139 | 2.63 | 4500 | 4.3991 |
| 4.1109 | 2.92 | 5000 | 4.3502 |
| 3.9085 | 3.21 | 5500 | 4.3470 |
| 3.8598 | 3.5 | 6000 | 4.3167 |
| 3.8525 | 3.79 | 6500 | 4.2818 |
| 3.7503 | 4.08 | 7000 | 4.2851 |
| 3.5747 | 4.38 | 7500 | 4.2769 |
| 3.5782 | 4.67 | 8000 | 4.2592 |
| 3.5679 | 4.96 | 8500 | 4.2398 |
| 3.3474 | 5.25 | 9000 | 4.2678 |
| 3.3278 | 5.54 | 9500 | 4.2623 |
| 3.3307 | 5.83 | 10000 | 4.2571 |
| 3.2522 | 6.13 | 10500 | 4.2674 |
| 3.1738 | 6.42 | 11000 | 4.2697 |
| 3.1687 | 6.71 | 11500 | 4.2692 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
