<!-- 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-base-no-repetition-no-cut
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3278
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 |
|---|---|---|---|
| 6.7605 | 0.31 | 500 | 5.6506 |
| 5.4119 | 0.62 | 1000 | 5.2199 |
| 5.0579 | 0.94 | 1500 | 4.9665 |
| 4.767 | 1.25 | 2000 | 4.8190 |
| 4.6276 | 1.56 | 2500 | 4.6923 |
| 4.5202 | 1.87 | 3000 | 4.5802 |
| 4.312 | 2.19 | 3500 | 4.5219 |
| 4.2135 | 2.5 | 4000 | 4.4518 |
| 4.1664 | 2.81 | 4500 | 4.3926 |
| 4.033 | 3.12 | 5000 | 4.3652 |
| 3.8843 | 3.44 | 5500 | 4.3407 |
| 3.8737 | 3.75 | 6000 | 4.3029 |
| 3.8047 | 4.06 | 6500 | 4.2883 |
| 3.5939 | 4.37 | 7000 | 4.2854 |
| 3.582 | 4.68 | 7500 | 4.2692 |
| 3.5745 | 5.0 | 8000 | 4.2540 |
| 3.3934 | 5.31 | 8500 | 4.2671 |
| 3.3874 | 5.62 | 9000 | 4.2653 |
| 3.3924 | 5.93 | 9500 | 4.2645 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
