<!-- 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. -->
gpt2-2_left_out_cbt
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.9359
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.9415 | 0.25 | 500 | 5.0484 |
| 4.7309 | 0.5 | 1000 | 4.6647 |
| 4.4299 | 0.75 | 1500 | 4.4406 |
| 4.2176 | 1.0 | 2000 | 4.3150 |
| 4.0185 | 1.25 | 2500 | 4.2114 |
| 3.9488 | 1.5 | 3000 | 4.1259 |
| 3.8765 | 1.75 | 3500 | 4.0526 |
| 3.8178 | 2.0 | 4000 | 4.0013 |
| 3.6252 | 2.25 | 4500 | 3.9677 |
| 3.6126 | 2.5 | 5000 | 3.9257 |
| 3.5877 | 2.75 | 5500 | 3.8858 |
| 3.5664 | 3.0 | 6000 | 3.8677 |
| 3.3588 | 3.25 | 6500 | 3.8655 |
| 3.3839 | 3.5 | 7000 | 3.8408 |
| 3.375 | 3.75 | 7500 | 3.8131 |
| 3.3702 | 4.0 | 8000 | 3.8096 |
| 3.1311 | 4.25 | 8500 | 3.8272 |
| 3.1718 | 4.5 | 9000 | 3.8071 |
| 3.1835 | 4.75 | 9500 | 3.7894 |
| 3.1802 | 5.01 | 10000 | 3.7920 |
| 2.9241 | 5.26 | 10500 | 3.8233 |
| 2.9597 | 5.51 | 11000 | 3.8156 |
| 2.9773 | 5.76 | 11500 | 3.8019 |
| 2.9708 | 6.01 | 12000 | 3.8077 |
| 2.7159 | 6.26 | 12500 | 3.8440 |
| 2.7495 | 6.51 | 13000 | 3.8459 |
| 2.761 | 6.76 | 13500 | 3.8406 |
| 2.7542 | 7.01 | 14000 | 3.8461 |
| 2.5238 | 7.26 | 14500 | 3.8832 |
| 2.5459 | 7.51 | 15000 | 3.8868 |
| 2.5638 | 7.76 | 15500 | 3.8872 |
| 2.555 | 8.01 | 16000 | 3.8932 |
| 2.388 | 8.26 | 16500 | 3.9161 |
| 2.4017 | 8.51 | 17000 | 3.9215 |
| 2.4056 | 8.76 | 17500 | 3.9236 |
| 2.3998 | 9.01 | 18000 | 3.9254 |
| 2.3199 | 9.26 | 18500 | 3.9339 |
| 2.3241 | 9.51 | 19000 | 3.9359 |
| 2.3205 | 9.76 | 19500 | 3.9359 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
