<!-- 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. -->
cl-norm-rarity-log-rarity-180k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.6308
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: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.6406 | 0.04 | 500 | 5.4417 |
| 5.2633 | 0.08 | 1000 | 5.0706 |
| 4.9048 | 0.13 | 1500 | 4.8767 |
| 4.663 | 0.17 | 2000 | 4.7614 |
| 4.4918 | 0.21 | 2500 | 4.6906 |
| 4.3581 | 0.25 | 3000 | 4.6319 |
| 4.2431 | 0.3 | 3500 | 4.6053 |
| 4.137 | 0.34 | 4000 | 4.5844 |
| 4.0418 | 0.38 | 4500 | 4.5624 |
| 3.9579 | 0.42 | 5000 | 4.5340 |
| 3.8642 | 0.46 | 5500 | 4.5333 |
| 3.7842 | 0.51 | 6000 | 4.5231 |
| 3.7016 | 0.55 | 6500 | 4.5284 |
| 3.6231 | 0.59 | 7000 | 4.5247 |
| 3.5474 | 0.63 | 7500 | 4.5221 |
| 3.4717 | 0.68 | 8000 | 4.5246 |
| 3.4052 | 0.72 | 8500 | 4.5312 |
| 3.3547 | 0.76 | 9000 | 4.5309 |
| 3.3033 | 0.8 | 9500 | 4.5291 |
| 3.2663 | 0.85 | 10000 | 4.5346 |
| 3.2378 | 0.89 | 10500 | 4.5343 |
| 3.2228 | 0.93 | 11000 | 4.5353 |
| 3.2127 | 0.97 | 11500 | 4.5352 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
