<!-- 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. -->
cbt-norm-rarity-log-rarity-no-cut
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.1028
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 |
|---|---|---|---|
| 6.3453 | 0.29 | 500 | 5.3315 |
| 5.033 | 0.58 | 1000 | 4.9200 |
| 4.6991 | 0.87 | 1500 | 4.6842 |
| 4.4349 | 1.16 | 2000 | 4.5403 |
| 4.2892 | 1.46 | 2500 | 4.4287 |
| 4.1929 | 1.75 | 3000 | 4.3281 |
| 4.0816 | 2.04 | 3500 | 4.2542 |
| 3.8848 | 2.33 | 4000 | 4.2078 |
| 3.8614 | 2.62 | 4500 | 4.1532 |
| 3.8318 | 2.91 | 5000 | 4.1052 |
| 3.6429 | 3.2 | 5500 | 4.0981 |
| 3.58 | 3.49 | 6000 | 4.0665 |
| 3.569 | 3.79 | 6500 | 4.0380 |
| 3.4854 | 4.08 | 7000 | 4.0323 |
| 3.3124 | 4.37 | 7500 | 4.0285 |
| 3.3128 | 4.66 | 8000 | 4.0149 |
| 3.2978 | 4.95 | 8500 | 4.0026 |
| 3.1549 | 5.24 | 9000 | 4.0129 |
| 3.1259 | 5.53 | 9500 | 4.0130 |
| 3.132 | 5.82 | 10000 | 4.0115 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
