<!-- 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-guten-rarity-all-7k-3k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1728
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 |
|---|---|---|---|
| 6.7108 | 0.3 | 500 | 5.6373 |
| 5.3452 | 0.6 | 1000 | 5.2026 |
| 4.9953 | 0.9 | 1500 | 4.9483 |
| 4.715 | 1.2 | 2000 | 4.7999 |
| 4.5638 | 1.5 | 2500 | 4.6688 |
| 4.4568 | 1.8 | 3000 | 4.5585 |
| 4.2928 | 2.1 | 3500 | 4.4953 |
| 4.1444 | 2.4 | 4000 | 4.4366 |
| 4.1046 | 2.7 | 4500 | 4.3829 |
| 4.0606 | 3.0 | 5000 | 4.3292 |
| 3.8079 | 3.3 | 5500 | 4.3257 |
| 3.8116 | 3.6 | 6000 | 4.2959 |
| 3.7882 | 3.9 | 6500 | 4.2612 |
| 3.6038 | 4.2 | 7000 | 4.2709 |
| 3.52 | 4.5 | 7500 | 4.2576 |
| 3.5139 | 4.8 | 8000 | 4.2421 |
| 3.4505 | 5.1 | 8500 | 4.2464 |
| 3.3277 | 5.4 | 9000 | 4.2481 |
| 3.3256 | 5.7 | 9500 | 4.2464 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
