<!-- 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-aochiles-14k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0042
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 |
|---|---|---|---|
| 6.7211 | 0.29 | 500 | 5.6349 |
| 5.3799 | 0.59 | 1000 | 5.1983 |
| 5.0235 | 0.88 | 1500 | 4.9554 |
| 4.7478 | 1.18 | 2000 | 4.8045 |
| 4.5979 | 1.47 | 2500 | 4.6854 |
| 4.4961 | 1.76 | 3000 | 4.5843 |
| 4.3569 | 2.06 | 3500 | 4.5164 |
| 4.1739 | 2.35 | 4000 | 4.4680 |
| 4.149 | 2.65 | 4500 | 4.4129 |
| 4.1093 | 2.94 | 5000 | 4.3581 |
| 3.8978 | 3.24 | 5500 | 4.3622 |
| 3.8629 | 3.53 | 6000 | 4.3327 |
| 3.8463 | 3.82 | 6500 | 4.3044 |
| 3.726 | 4.12 | 7000 | 4.3127 |
| 3.5714 | 4.41 | 7500 | 4.3116 |
| 3.5846 | 4.71 | 8000 | 4.2872 |
| 3.5668 | 5.0 | 8500 | 4.2693 |
| 3.3167 | 5.29 | 9000 | 4.3073 |
| 3.3274 | 5.59 | 9500 | 4.3060 |
| 3.3202 | 5.88 | 10000 | 4.3010 |
| 3.2207 | 6.18 | 10500 | 4.3137 |
| 3.1707 | 6.47 | 11000 | 4.3147 |
| 3.1663 | 6.76 | 11500 | 4.3166 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
