metadata
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: gpt2-concat-mod-datasets-rarity1
results: []
gpt2-concat-mod-datasets-rarity1
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 3.0210
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: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 6.7299 | 0.3 | 500 | 5.6367 |
| 5.3814 | 0.59 | 1000 | 5.2097 |
| 5.0305 | 0.89 | 1500 | 4.9565 |
| 4.7532 | 1.18 | 2000 | 4.8178 |
| 4.6062 | 1.48 | 2500 | 4.6913 |
| 4.4987 | 1.78 | 3000 | 4.5883 |
| 4.3593 | 2.07 | 3500 | 4.5246 |
| 4.1845 | 2.37 | 4000 | 4.4796 |
| 4.1539 | 2.66 | 4500 | 4.4191 |
| 4.1258 | 2.96 | 5000 | 4.3681 |
| 3.898 | 3.26 | 5500 | 4.3751 |
| 3.8758 | 3.55 | 6000 | 4.3495 |
| 3.8598 | 3.85 | 6500 | 4.3088 |
| 3.7173 | 4.14 | 7000 | 4.3340 |
| 3.5968 | 4.44 | 7500 | 4.3170 |
| 3.5934 | 4.74 | 8000 | 4.3049 |
| 3.5491 | 5.03 | 8500 | 4.3103 |
| 3.3358 | 5.33 | 9000 | 4.3192 |
| 3.3363 | 5.62 | 9500 | 4.3181 |
| 3.3409 | 5.92 | 10000 | 4.3105 |
| 3.2189 | 6.22 | 10500 | 4.3290 |
| 3.1812 | 6.51 | 11000 | 4.3286 |
| 3.1879 | 6.81 | 11500 | 4.3297 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3