<!-- 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. -->
dp-guten-rarity-all-end-2p5k-ctx-256
This model is a fine-tuned version of gpt2 on the generator dataset. It achieves the following results on the evaluation set:
- Loss: 4.2315
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 |
|---|---|---|---|
| 7.8026 | 0.21 | 200 | 6.2765 |
| 5.9069 | 0.43 | 400 | 5.7198 |
| 5.4957 | 0.64 | 600 | 5.4158 |
| 5.2157 | 0.86 | 800 | 5.1901 |
| 4.9843 | 1.07 | 1000 | 5.0506 |
| 4.8053 | 1.29 | 1200 | 4.9027 |
| 4.6827 | 1.5 | 1400 | 4.7836 |
| 4.5784 | 1.72 | 1600 | 4.6794 |
| 4.4881 | 1.93 | 1800 | 4.5957 |
| 4.2837 | 2.15 | 2000 | 4.5246 |
| 4.213 | 2.36 | 2200 | 4.4665 |
| 4.165 | 2.58 | 2400 | 4.4076 |
| 4.1205 | 2.79 | 2600 | 4.3531 |
| 4.0806 | 3.01 | 2800 | 4.3271 |
| 3.8411 | 3.22 | 3000 | 4.3015 |
| 3.8429 | 3.44 | 3200 | 4.2754 |
| 3.8312 | 3.65 | 3400 | 4.2507 |
| 3.8174 | 3.87 | 3600 | 4.2153 |
| 3.7124 | 4.08 | 3800 | 4.2111 |
| 3.565 | 4.3 | 4000 | 4.2049 |
| 3.5597 | 4.51 | 4200 | 4.1889 |
| 3.5384 | 4.73 | 4400 | 4.1749 |
| 3.5301 | 4.94 | 4600 | 4.1637 |
| 3.4106 | 5.16 | 4800 | 4.1717 |
| 3.3593 | 5.37 | 5000 | 4.1683 |
| 3.356 | 5.59 | 5200 | 4.1673 |
| 3.356 | 5.8 | 5400 | 4.1657 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.11.0+cu113
- Datasets 2.13.0
- Tokenizers 0.13.3
