<!-- 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. -->
10032023
This model is a fine-tuned version of bigcode/santacoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2642
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: 1
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 100
- training_steps: 4000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.6282 | 0.05 | 200 | 0.4105 |
| 1.7635 | 0.1 | 400 | 0.5228 |
| 1.7029 | 0.15 | 600 | 0.8193 |
| 1.6817 | 0.2 | 800 | 1.6320 |
| 1.6822 | 0.25 | 1000 | 2.8463 |
| 1.671 | 0.3 | 1200 | 3.4860 |
| 1.6698 | 0.35 | 1400 | 4.1775 |
| 1.6631 | 0.4 | 1600 | 5.2973 |
| 1.663 | 0.45 | 1800 | 5.8655 |
| 1.6599 | 0.5 | 2000 | 5.8967 |
| 1.6595 | 0.55 | 2200 | 0.2873 |
| 1.6586 | 0.6 | 2400 | 0.3041 |
| 1.6564 | 0.65 | 2600 | 0.3210 |
| 1.658 | 0.7 | 2800 | 0.3262 |
| 1.6549 | 0.75 | 3000 | 0.3136 |
| 1.6498 | 0.8 | 3200 | 0.3232 |
| 1.6462 | 0.85 | 3400 | 0.3195 |
| 1.6454 | 0.9 | 3600 | 0.3216 |
| 0.2173 | 0.95 | 3800 | 0.2726 |
| 1.6619 | 1.0 | 4000 | 0.2642 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
