<!-- 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. -->
all_base_rarity_neg_log_rarity_rev_no_shuffle
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.8763
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 |
|---|---|---|---|
| 6.493 | 0.31 | 500 | 5.4634 |
| 5.2061 | 0.62 | 1000 | 5.1283 |
| 4.8646 | 0.94 | 1500 | 4.9720 |
| 4.6091 | 1.25 | 2000 | 4.8902 |
| 4.4895 | 1.56 | 2500 | 4.8280 |
| 4.3953 | 1.87 | 3000 | 4.7810 |
| 4.2229 | 2.19 | 3500 | 4.7672 |
| 4.1396 | 2.5 | 4000 | 4.7502 |
| 4.1055 | 2.81 | 4500 | 4.7302 |
| 3.9813 | 3.12 | 5000 | 4.7672 |
| 3.8461 | 3.44 | 5500 | 4.7478 |
| 3.8342 | 3.75 | 6000 | 4.7348 |
| 3.7637 | 4.06 | 6500 | 4.7609 |
| 3.5734 | 4.37 | 7000 | 4.7842 |
| 3.5696 | 4.68 | 7500 | 4.7787 |
| 3.549 | 5.0 | 8000 | 4.7897 |
| 3.3841 | 5.31 | 8500 | 4.8205 |
| 3.3813 | 5.62 | 9000 | 4.8209 |
| 3.3816 | 5.93 | 9500 | 4.8245 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
