<!-- 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-concat-all-rarity-all-29k-3k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.2385
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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7809 | 0.32 | 500 | 5.8462 |
| 5.4862 | 0.64 | 1000 | 5.4612 |
| 5.1383 | 0.97 | 1500 | 5.2431 |
| 4.8617 | 1.29 | 2000 | 5.1499 |
| 4.7488 | 1.61 | 2500 | 5.0779 |
| 4.6532 | 1.93 | 3000 | 4.9910 |
| 4.4482 | 2.26 | 3500 | 4.9776 |
| 4.3986 | 2.58 | 4000 | 4.9341 |
| 4.3585 | 2.9 | 4500 | 4.9077 |
| 4.1722 | 3.22 | 5000 | 4.9013 |
| 4.1059 | 3.55 | 5500 | 4.8956 |
| 4.0999 | 3.87 | 6000 | 4.8572 |
| 3.9186 | 4.19 | 6500 | 4.8850 |
| 3.8238 | 4.51 | 7000 | 4.8874 |
| 3.8274 | 4.84 | 7500 | 4.8707 |
| 3.688 | 5.16 | 8000 | 4.9006 |
| 3.5678 | 5.48 | 8500 | 4.9129 |
| 3.5663 | 5.8 | 9000 | 4.9066 |
| 3.4981 | 6.13 | 9500 | 4.9232 |
| 3.4058 | 6.45 | 10000 | 4.9276 |
| 3.4052 | 6.77 | 10500 | 4.9278 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
