<!-- 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. -->
gpt2-3-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: 3.9275
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 |
|---|---|---|---|
| 5.9917 | 0.24 | 500 | 5.0580 |
| 4.751 | 0.48 | 1000 | 4.6761 |
| 4.4491 | 0.72 | 1500 | 4.4474 |
| 4.2552 | 0.96 | 2000 | 4.3018 |
| 4.0564 | 1.21 | 2500 | 4.2130 |
| 3.9694 | 1.45 | 3000 | 4.1354 |
| 3.9064 | 1.69 | 3500 | 4.0597 |
| 3.8419 | 1.93 | 4000 | 3.9915 |
| 3.6722 | 2.17 | 4500 | 3.9682 |
| 3.6318 | 2.41 | 5000 | 3.9315 |
| 3.6106 | 2.65 | 5500 | 3.8886 |
| 3.5928 | 2.89 | 6000 | 3.8514 |
| 3.4548 | 3.13 | 6500 | 3.8612 |
| 3.3861 | 3.38 | 7000 | 3.8411 |
| 3.393 | 3.62 | 7500 | 3.8154 |
| 3.3954 | 3.86 | 8000 | 3.7894 |
| 3.2757 | 4.1 | 8500 | 3.8165 |
| 3.1711 | 4.34 | 9000 | 3.8133 |
| 3.196 | 4.58 | 9500 | 3.7968 |
| 3.1968 | 4.82 | 10000 | 3.7750 |
| 3.1316 | 5.06 | 10500 | 3.8042 |
| 2.9476 | 5.3 | 11000 | 3.8150 |
| 2.9825 | 5.54 | 11500 | 3.8057 |
| 2.9945 | 5.79 | 12000 | 3.7922 |
| 2.9682 | 6.03 | 12500 | 3.8095 |
| 2.7376 | 6.27 | 13000 | 3.8392 |
| 2.7689 | 6.51 | 13500 | 3.8374 |
| 2.78 | 6.75 | 14000 | 3.8313 |
| 2.7801 | 6.99 | 14500 | 3.8215 |
| 2.5564 | 7.23 | 15000 | 3.8731 |
| 2.5648 | 7.47 | 15500 | 3.8790 |
| 2.5779 | 7.71 | 16000 | 3.8779 |
| 2.5815 | 7.96 | 16500 | 3.8749 |
| 2.4329 | 8.2 | 17000 | 3.9075 |
| 2.4187 | 8.44 | 17500 | 3.9123 |
| 2.4313 | 8.68 | 18000 | 3.9145 |
| 2.4232 | 8.92 | 18500 | 3.9151 |
| 2.3723 | 9.16 | 19000 | 3.9246 |
| 2.3473 | 9.4 | 19500 | 3.9267 |
| 2.3464 | 9.64 | 20000 | 3.9275 |
| 2.3445 | 9.88 | 20500 | 3.9275 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
