<!-- 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-aochildes-len-16k-rarity-all-no-self-4k-1p2k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.1934
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 |
|---|---|---|---|
| 6.7394 | 0.3 | 500 | 5.6331 |
| 5.3748 | 0.59 | 1000 | 5.2044 |
| 5.0309 | 0.89 | 1500 | 4.9493 |
| 4.7518 | 1.18 | 2000 | 4.8041 |
| 4.5959 | 1.48 | 2500 | 4.6818 |
| 4.4873 | 1.77 | 3000 | 4.5795 |
| 4.3537 | 2.07 | 3500 | 4.5123 |
| 4.1676 | 2.36 | 4000 | 4.4632 |
| 4.1387 | 2.66 | 4500 | 4.3957 |
| 4.0998 | 2.95 | 5000 | 4.3479 |
| 3.8663 | 3.25 | 5500 | 4.3465 |
| 3.8329 | 3.54 | 6000 | 4.3101 |
| 3.8222 | 3.84 | 6500 | 4.2757 |
| 3.6816 | 4.13 | 7000 | 4.2834 |
| 3.5463 | 4.43 | 7500 | 4.2723 |
| 3.5397 | 4.72 | 8000 | 4.2563 |
| 3.5124 | 5.02 | 8500 | 4.2552 |
| 3.3501 | 5.31 | 9000 | 4.2619 |
| 3.3456 | 5.61 | 9500 | 4.2600 |
| 3.3437 | 5.9 | 10000 | 4.2593 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
