gpt2_small_baby_10M_32768_42
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 4.3856
- Accuracy: 0.3014
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: 0.0001
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 40000
- training_steps: 100000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|
| 5.0453 | 2.2 | 2000 | 6.0769 | 0.1990 |
| 4.111 | 4.4 | 4000 | 5.4544 | 0.2166 |
| 3.8617 | 6.59 | 6000 | 5.1646 | 0.2352 |
| 3.5936 | 8.79 | 8000 | 4.9026 | 0.2611 |
| 3.4285 | 10.99 | 10000 | 4.7411 | 0.2738 |
| 3.3107 | 13.19 | 12000 | 4.5934 | 0.2839 |
| 3.2153 | 15.38 | 14000 | 4.4873 | 0.2929 |
| 3.1348 | 17.58 | 16000 | 4.4476 | 0.2973 |
| 3.0503 | 19.78 | 18000 | 4.4134 | 0.3005 |
| 2.978 | 21.98 | 20000 | 4.3856 | 0.3014 |
| 2.8939 | 24.18 | 22000 | 4.4394 | 0.3004 |
| 2.8208 | 26.37 | 24000 | 4.4451 | 0.3011 |
| 2.742 | 28.57 | 26000 | 4.4888 | 0.2987 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.9.0+cu128
- Datasets 4.1.1
- Tokenizers 0.13.3
