<!-- 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-miss-qed-seed
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.1573
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 |
---|---|---|---|
6.3409 | 0.32 | 500 | 5.3423 |
5.0125 | 0.65 | 1000 | 4.9233 |
4.6826 | 0.97 | 1500 | 4.6888 |
4.3996 | 1.3 | 2000 | 4.5489 |
4.2736 | 1.62 | 2500 | 4.4300 |
4.1603 | 1.95 | 3000 | 4.3281 |
3.9399 | 2.27 | 3500 | 4.2800 |
3.8891 | 2.59 | 4000 | 4.2130 |
3.8439 | 2.92 | 4500 | 4.1511 |
3.6447 | 3.24 | 5000 | 4.1418 |
3.5886 | 3.57 | 5500 | 4.1108 |
3.5658 | 3.89 | 6000 | 4.0761 |
3.3856 | 4.22 | 6500 | 4.0825 |
3.314 | 4.54 | 7000 | 4.0683 |
3.3035 | 4.86 | 7500 | 4.0534 |
3.1883 | 5.19 | 8000 | 4.0608 |
3.1227 | 5.51 | 8500 | 4.0601 |
3.1192 | 5.84 | 9000 | 4.0592 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3