metadata
license: mit
tags:
- generated_from_trainer
datasets:
- generator
model-index:
- name: switchboard-rarity-seed
results: []
switchboard-rarity-seed
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.0985
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 |
---|---|---|---|
6.3581 | 0.29 | 500 | 5.3466 |
5.0332 | 0.58 | 1000 | 4.9336 |
4.7065 | 0.87 | 1500 | 4.6924 |
4.4439 | 1.17 | 2000 | 4.5465 |
4.2929 | 1.46 | 2500 | 4.4328 |
4.1869 | 1.75 | 3000 | 4.3248 |
4.0802 | 2.04 | 3500 | 4.2481 |
3.8877 | 2.33 | 4000 | 4.2060 |
3.8547 | 2.62 | 4500 | 4.1542 |
3.83 | 2.92 | 5000 | 4.0982 |
3.6375 | 3.21 | 5500 | 4.0946 |
3.5896 | 3.5 | 6000 | 4.0648 |
3.5596 | 3.79 | 6500 | 4.0309 |
3.474 | 4.08 | 7000 | 4.0282 |
3.3101 | 4.37 | 7500 | 4.0247 |
3.3055 | 4.66 | 8000 | 4.0122 |
3.2891 | 4.96 | 8500 | 3.9981 |
3.1562 | 5.25 | 9000 | 4.0102 |
3.1289 | 5.54 | 9500 | 4.0093 |
3.1216 | 5.83 | 10000 | 4.0085 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3