<!-- 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. -->
all-indv-norm-rarity-log-rarity
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.7065
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 |
|---|---|---|---|
| 6.4518 | 0.31 | 500 | 5.3977 |
| 5.1596 | 0.62 | 1000 | 5.0441 |
| 4.8387 | 0.93 | 1500 | 4.8717 |
| 4.5792 | 1.25 | 2000 | 4.7824 |
| 4.4634 | 1.56 | 2500 | 4.6973 |
| 4.3707 | 1.87 | 3000 | 4.6444 |
| 4.2028 | 2.18 | 3500 | 4.6394 |
| 4.1179 | 2.49 | 4000 | 4.6145 |
| 4.0765 | 2.8 | 4500 | 4.5740 |
| 3.9582 | 3.11 | 5000 | 4.6031 |
| 3.821 | 3.42 | 5500 | 4.5809 |
| 3.807 | 3.74 | 6000 | 4.5636 |
| 3.7537 | 4.05 | 6500 | 4.5897 |
| 3.55 | 4.36 | 7000 | 4.5981 |
| 3.5465 | 4.67 | 7500 | 4.6095 |
| 3.5329 | 4.98 | 8000 | 4.5984 |
| 3.3775 | 5.29 | 8500 | 4.6392 |
| 3.3671 | 5.6 | 9000 | 4.6434 |
| 3.3627 | 5.92 | 9500 | 4.6471 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
