<!-- 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-rarity-neg-log-rarity-end-p8k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.1023
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 |
|---|---|---|---|
| 6.3445 | 0.29 | 500 | 5.3418 |
| 5.0296 | 0.58 | 1000 | 4.9281 |
| 4.7069 | 0.88 | 1500 | 4.6922 |
| 4.4415 | 1.17 | 2000 | 4.5462 |
| 4.2976 | 1.46 | 2500 | 4.4280 |
| 4.1934 | 1.75 | 3000 | 4.3279 |
| 4.0772 | 2.04 | 3500 | 4.2526 |
| 3.8904 | 2.33 | 4000 | 4.2097 |
| 3.8664 | 2.63 | 4500 | 4.1528 |
| 3.8245 | 2.92 | 5000 | 4.1044 |
| 3.6329 | 3.21 | 5500 | 4.0963 |
| 3.578 | 3.5 | 6000 | 4.0636 |
| 3.5675 | 3.79 | 6500 | 4.0358 |
| 3.4841 | 4.08 | 7000 | 4.0324 |
| 3.3109 | 4.38 | 7500 | 4.0293 |
| 3.314 | 4.67 | 8000 | 4.0142 |
| 3.2984 | 4.96 | 8500 | 4.0014 |
| 3.155 | 5.25 | 9000 | 4.0138 |
| 3.1322 | 5.54 | 9500 | 4.0123 |
| 3.1327 | 5.83 | 10000 | 4.0118 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
