<!-- 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. -->
MammoLLM_bz32_acc4_lr1e4_large
This model is a fine-tuned version of gpt2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1323
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: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 1000
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.5563 | 0.48 | 500 | 2.1117 |
| 1.963 | 0.96 | 1000 | 1.6979 |
| 1.6708 | 1.44 | 1500 | 1.5274 |
| 1.5456 | 1.92 | 2000 | 1.4406 |
| 1.452 | 2.4 | 2500 | 1.3906 |
| 1.4081 | 2.87 | 3000 | 1.3467 |
| 1.346 | 3.35 | 3500 | 1.3180 |
| 1.3133 | 3.83 | 4000 | 1.2879 |
| 1.2622 | 4.31 | 4500 | 1.2587 |
| 1.2337 | 4.79 | 5000 | 1.2409 |
| 1.1872 | 5.27 | 5500 | 1.2200 |
| 1.158 | 5.75 | 6000 | 1.2001 |
| 1.12 | 6.23 | 6500 | 1.1842 |
| 1.0878 | 6.71 | 7000 | 1.1682 |
| 1.0622 | 7.19 | 7500 | 1.1585 |
| 1.0307 | 7.66 | 8000 | 1.1475 |
| 1.015 | 8.14 | 8500 | 1.1406 |
| 0.991 | 8.62 | 9000 | 1.1355 |
| 0.9858 | 9.1 | 9500 | 1.1336 |
| 0.9718 | 9.58 | 10000 | 1.1323 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.3
- Tokenizers 0.13.3
