<!-- 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. -->
gpt2-2-og-concat-modified-aochild
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.9262
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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.9891 | 0.24 | 500 | 5.0538 |
| 4.7513 | 0.48 | 1000 | 4.6760 |
| 4.4523 | 0.72 | 1500 | 4.4485 |
| 4.2602 | 0.96 | 2000 | 4.3053 |
| 4.0617 | 1.21 | 2500 | 4.2166 |
| 3.9742 | 1.45 | 3000 | 4.1365 |
| 3.9095 | 1.69 | 3500 | 4.0632 |
| 3.8462 | 1.93 | 4000 | 3.9949 |
| 3.6761 | 2.17 | 4500 | 3.9718 |
| 3.6346 | 2.41 | 5000 | 3.9336 |
| 3.613 | 2.65 | 5500 | 3.8883 |
| 3.5949 | 2.89 | 6000 | 3.8502 |
| 3.4561 | 3.13 | 6500 | 3.8626 |
| 3.387 | 3.38 | 7000 | 3.8393 |
| 3.3931 | 3.62 | 7500 | 3.8152 |
| 3.395 | 3.86 | 8000 | 3.7882 |
| 3.2751 | 4.1 | 8500 | 3.8162 |
| 3.1697 | 4.34 | 9000 | 3.8117 |
| 3.1949 | 4.58 | 9500 | 3.7952 |
| 3.1957 | 4.82 | 10000 | 3.7726 |
| 3.1301 | 5.06 | 10500 | 3.8013 |
| 2.9449 | 5.3 | 11000 | 3.8132 |
| 2.9803 | 5.54 | 11500 | 3.8048 |
| 2.9921 | 5.79 | 12000 | 3.7903 |
| 2.9654 | 6.03 | 12500 | 3.8054 |
| 2.7336 | 6.27 | 13000 | 3.8363 |
| 2.7653 | 6.51 | 13500 | 3.8379 |
| 2.7754 | 6.75 | 14000 | 3.8285 |
| 2.777 | 6.99 | 14500 | 3.8186 |
| 2.5506 | 7.23 | 15000 | 3.8731 |
| 2.5598 | 7.47 | 15500 | 3.8769 |
| 2.5731 | 7.71 | 16000 | 3.8768 |
| 2.5762 | 7.96 | 16500 | 3.8744 |
| 2.4267 | 8.2 | 17000 | 3.9055 |
| 2.4121 | 8.44 | 17500 | 3.9110 |
| 2.4249 | 8.68 | 18000 | 3.9133 |
| 2.4157 | 8.92 | 18500 | 3.9140 |
| 2.366 | 9.16 | 19000 | 3.9237 |
| 2.3398 | 9.4 | 19500 | 3.9252 |
| 2.3398 | 9.64 | 20000 | 3.9263 |
| 2.3365 | 9.88 | 20500 | 3.9262 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
