<!-- 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. -->
gp2-concat-guten-mod-rm-2p3k-rarity-all-5k-p22k
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.3210
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 |
---|---|---|---|
6.6886 | 0.3 | 500 | 5.6426 |
5.3235 | 0.59 | 1000 | 5.2038 |
4.9823 | 0.89 | 1500 | 4.9471 |
4.7179 | 1.18 | 2000 | 4.8049 |
4.5528 | 1.48 | 2500 | 4.6805 |
4.4405 | 1.77 | 3000 | 4.5719 |
4.3068 | 2.07 | 3500 | 4.5048 |
4.1286 | 2.36 | 4000 | 4.4539 |
4.1002 | 2.66 | 4500 | 4.3893 |
4.0576 | 2.95 | 5000 | 4.3358 |
3.8317 | 3.25 | 5500 | 4.3369 |
3.7985 | 3.54 | 6000 | 4.3073 |
3.7837 | 3.84 | 6500 | 4.2695 |
3.6537 | 4.13 | 7000 | 4.2751 |
3.5154 | 4.43 | 7500 | 4.2636 |
3.5075 | 4.72 | 8000 | 4.2498 |
3.4858 | 5.02 | 8500 | 4.2444 |
3.3198 | 5.31 | 9000 | 4.2521 |
3.3198 | 5.61 | 9500 | 4.2505 |
3.3174 | 5.9 | 10000 | 4.2500 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3