<!-- 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. -->
bnc-rarity-no-cut-shuffled
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3207
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 |
|---|---|---|---|
| 6.7157 | 0.29 | 500 | 5.6437 |
| 5.3513 | 0.58 | 1000 | 5.2021 |
| 5.0016 | 0.88 | 1500 | 4.9595 |
| 4.7286 | 1.17 | 2000 | 4.8122 |
| 4.5693 | 1.46 | 2500 | 4.6857 |
| 4.4647 | 1.75 | 3000 | 4.5770 |
| 4.3308 | 2.05 | 3500 | 4.5068 |
| 4.1402 | 2.34 | 4000 | 4.4574 |
| 4.1123 | 2.63 | 4500 | 4.3983 |
| 4.0711 | 2.92 | 5000 | 4.3468 |
| 3.8657 | 3.22 | 5500 | 4.3414 |
| 3.8086 | 3.51 | 6000 | 4.3099 |
| 3.7977 | 3.8 | 6500 | 4.2728 |
| 3.6947 | 4.09 | 7000 | 4.2729 |
| 3.5188 | 4.39 | 7500 | 4.2684 |
| 3.5211 | 4.68 | 8000 | 4.2523 |
| 3.5159 | 4.97 | 8500 | 4.2387 |
| 3.3414 | 5.26 | 9000 | 4.2532 |
| 3.3357 | 5.56 | 9500 | 4.2520 |
| 3.328 | 5.85 | 10000 | 4.2517 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
