<!-- 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-guten-not-merged
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.0326
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 |
---|---|---|---|
6.3581 | 0.29 | 500 | 5.3149 |
5.0366 | 0.58 | 1000 | 4.9005 |
4.6975 | 0.87 | 1500 | 4.6545 |
4.4485 | 1.16 | 2000 | 4.5117 |
4.289 | 1.45 | 2500 | 4.3912 |
4.1875 | 1.74 | 3000 | 4.2850 |
4.084 | 2.02 | 3500 | 4.2030 |
3.8847 | 2.31 | 4000 | 4.1661 |
3.8566 | 2.6 | 4500 | 4.1083 |
3.8167 | 2.89 | 5000 | 4.0563 |
3.6493 | 3.18 | 5500 | 4.0476 |
3.5729 | 3.47 | 6000 | 4.0204 |
3.5597 | 3.76 | 6500 | 3.9876 |
3.4955 | 4.05 | 7000 | 3.9750 |
3.3056 | 4.34 | 7500 | 3.9767 |
3.2984 | 4.63 | 8000 | 3.9632 |
3.2904 | 4.92 | 8500 | 3.9508 |
3.1679 | 5.21 | 9000 | 3.9591 |
3.1169 | 5.49 | 9500 | 3.9585 |
3.1184 | 5.78 | 10000 | 3.9574 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3