metadata
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: aochildes-rarity
results: []
aochildes-rarity
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.0570
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 |
|---|---|---|---|
| 6.3684 | 0.29 | 500 | 5.3161 |
| 5.0527 | 0.58 | 1000 | 4.8912 |
| 4.716 | 0.87 | 1500 | 4.6588 |
| 4.4472 | 1.16 | 2000 | 4.5116 |
| 4.3056 | 1.46 | 2500 | 4.3936 |
| 4.2065 | 1.75 | 3000 | 4.3008 |
| 4.0872 | 2.04 | 3500 | 4.2252 |
| 3.8993 | 2.33 | 4000 | 4.1825 |
| 3.8715 | 2.62 | 4500 | 4.1240 |
| 3.8388 | 2.91 | 5000 | 4.0762 |
| 3.6552 | 3.2 | 5500 | 4.0680 |
| 3.592 | 3.49 | 6000 | 4.0385 |
| 3.5701 | 3.79 | 6500 | 4.0111 |
| 3.4953 | 4.08 | 7000 | 4.0047 |
| 3.3186 | 4.37 | 7500 | 4.0017 |
| 3.3174 | 4.66 | 8000 | 3.9888 |
| 3.3059 | 4.95 | 8500 | 3.9759 |
| 3.1594 | 5.24 | 9000 | 3.9886 |
| 3.1356 | 5.53 | 9500 | 3.9881 |
| 3.1353 | 5.82 | 10000 | 3.9870 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3