<!-- 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-cbt-rarity-2k-p3k-rerun
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0111
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 |
|---|---|---|---|
| 6.7193 | 0.29 | 500 | 5.6348 |
| 5.3717 | 0.58 | 1000 | 5.2023 |
| 5.0316 | 0.87 | 1500 | 4.9479 |
| 4.7526 | 1.17 | 2000 | 4.8054 |
| 4.5908 | 1.46 | 2500 | 4.6881 |
| 4.4855 | 1.75 | 3000 | 4.5743 |
| 4.37 | 2.04 | 3500 | 4.4982 |
| 4.1665 | 2.33 | 4000 | 4.4536 |
| 4.1433 | 2.62 | 4500 | 4.3994 |
| 4.1049 | 2.91 | 5000 | 4.3436 |
| 3.9129 | 3.21 | 5500 | 4.3424 |
| 3.8605 | 3.5 | 6000 | 4.3136 |
| 3.8444 | 3.79 | 6500 | 4.2788 |
| 3.7551 | 4.08 | 7000 | 4.2807 |
| 3.5722 | 4.37 | 7500 | 4.2752 |
| 3.58 | 4.66 | 8000 | 4.2599 |
| 3.5691 | 4.95 | 8500 | 4.2378 |
| 3.3607 | 5.24 | 9000 | 4.2659 |
| 3.3218 | 5.54 | 9500 | 4.2623 |
| 3.3249 | 5.83 | 10000 | 4.2549 |
| 3.2621 | 6.12 | 10500 | 4.2656 |
| 3.1746 | 6.41 | 11000 | 4.2682 |
| 3.1689 | 6.7 | 11500 | 4.2679 |
| 3.17 | 6.99 | 12000 | 4.2677 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
