<!-- 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-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: 3.9256
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 |
|---|---|---|---|
| 5.9858 | 0.24 | 500 | 5.0593 |
| 4.752 | 0.48 | 1000 | 4.6760 |
| 4.4497 | 0.72 | 1500 | 4.4435 |
| 4.2543 | 0.96 | 2000 | 4.2976 |
| 4.0555 | 1.21 | 2500 | 4.2137 |
| 3.9693 | 1.45 | 3000 | 4.1335 |
| 3.906 | 1.69 | 3500 | 4.0568 |
| 3.8429 | 1.93 | 4000 | 3.9920 |
| 3.6732 | 2.17 | 4500 | 3.9691 |
| 3.6327 | 2.41 | 5000 | 3.9306 |
| 3.6116 | 2.65 | 5500 | 3.8914 |
| 3.5938 | 2.89 | 6000 | 3.8513 |
| 3.455 | 3.13 | 6500 | 3.8610 |
| 3.3859 | 3.38 | 7000 | 3.8405 |
| 3.3923 | 3.62 | 7500 | 3.8156 |
| 3.3951 | 3.86 | 8000 | 3.7887 |
| 3.2753 | 4.1 | 8500 | 3.8143 |
| 3.1704 | 4.34 | 9000 | 3.8108 |
| 3.1945 | 4.58 | 9500 | 3.7931 |
| 3.1957 | 4.82 | 10000 | 3.7730 |
| 3.1308 | 5.06 | 10500 | 3.7997 |
| 2.9454 | 5.3 | 11000 | 3.8140 |
| 2.981 | 5.54 | 11500 | 3.8037 |
| 2.9917 | 5.79 | 12000 | 3.7886 |
| 2.9661 | 6.03 | 12500 | 3.8061 |
| 2.7333 | 6.27 | 13000 | 3.8368 |
| 2.7658 | 6.51 | 13500 | 3.8365 |
| 2.7757 | 6.75 | 14000 | 3.8304 |
| 2.7771 | 6.99 | 14500 | 3.8187 |
| 2.5518 | 7.23 | 15000 | 3.8726 |
| 2.56 | 7.47 | 15500 | 3.8759 |
| 2.5737 | 7.71 | 16000 | 3.8764 |
| 2.5772 | 7.96 | 16500 | 3.8738 |
| 2.4267 | 8.2 | 17000 | 3.9046 |
| 2.4129 | 8.44 | 17500 | 3.9102 |
| 2.4256 | 8.68 | 18000 | 3.9135 |
| 2.4177 | 8.92 | 18500 | 3.9138 |
| 2.3675 | 9.16 | 19000 | 3.9222 |
| 2.3412 | 9.4 | 19500 | 3.9246 |
| 2.3399 | 9.64 | 20000 | 3.9256 |
| 2.3381 | 9.88 | 20500 | 3.9256 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
