<!-- 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. -->
santacoder-finetuned-the-stack-dockerfiles
This model is a fine-tuned version of bigcode/santacoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8741
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: 5e-05
- train_batch_size: 2
- eval_batch_size: 2
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 8
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 10000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.3691 | 0.05 | 500 | 1.2915 |
| 1.267 | 0.1 | 1000 | 1.1962 |
| 1.1998 | 0.15 | 1500 | 1.1459 |
| 1.0753 | 0.2 | 2000 | 1.1130 |
| 1.0762 | 0.25 | 2500 | 1.0797 |
| 1.1222 | 0.3 | 3000 | 1.0532 |
| 1.0542 | 0.35 | 3500 | 1.0282 |
| 1.0853 | 0.4 | 4000 | 1.0050 |
| 1.0619 | 0.45 | 4500 | 0.9803 |
| 1.0484 | 0.5 | 5000 | 0.9587 |
| 0.9795 | 0.55 | 5500 | 0.9501 |
| 0.9439 | 0.6 | 6000 | 0.9294 |
| 0.9642 | 0.65 | 6500 | 0.9143 |
| 0.935 | 0.7 | 7000 | 0.9028 |
| 0.836 | 0.75 | 7500 | 0.8936 |
| 0.9128 | 0.8 | 8000 | 0.8844 |
| 0.9286 | 0.85 | 8500 | 0.8795 |
| 0.9899 | 0.9 | 9000 | 0.8758 |
| 0.8307 | 0.95 | 9500 | 0.8745 |
| 0.9366 | 1.0 | 10000 | 0.8741 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.7.1
- Tokenizers 0.13.2
