|
--- |
|
license: bigcode-openrail-m |
|
base_model: bigcode/santacoder |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: SCoder-APPS |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# SCoder-APPS |
|
|
|
This model is a fine-tuned version of [bigcode/santacoder](https://huggingface.co/bigcode/santacoder) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.8114 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 4 |
|
- total_train_batch_size: 16 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- training_steps: 5000 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:----:|:---------------:| |
|
| 1.006 | 0.04 | 200 | 1.0234 | |
|
| 0.9936 | 0.08 | 400 | 0.9176 | |
|
| 0.9287 | 0.12 | 600 | 0.9170 | |
|
| 0.8434 | 0.16 | 800 | 0.8872 | |
|
| 0.8223 | 0.2 | 1000 | 0.8750 | |
|
| 0.8129 | 0.24 | 1200 | 0.8720 | |
|
| 0.8612 | 0.28 | 1400 | 0.8624 | |
|
| 0.777 | 0.32 | 1600 | 0.8426 | |
|
| 0.7444 | 0.36 | 1800 | 0.8453 | |
|
| 0.6214 | 0.4 | 2000 | 0.8428 | |
|
| 0.6856 | 0.44 | 2200 | 0.8365 | |
|
| 0.6463 | 0.48 | 2400 | 0.8379 | |
|
| 0.5872 | 0.52 | 2600 | 0.8226 | |
|
| 0.6271 | 0.56 | 2800 | 0.8132 | |
|
| 0.5772 | 0.6 | 3000 | 0.8237 | |
|
| 0.568 | 0.64 | 3200 | 0.8097 | |
|
| 0.5718 | 0.68 | 3400 | 0.8025 | |
|
| 0.5407 | 0.72 | 3600 | 0.8222 | |
|
| 0.4531 | 0.76 | 3800 | 0.8164 | |
|
| 0.5571 | 0.8 | 4000 | 0.8209 | |
|
| 0.4933 | 0.84 | 4200 | 0.8218 | |
|
| 0.4749 | 0.88 | 4400 | 0.8176 | |
|
| 0.4907 | 0.92 | 4600 | 0.8137 | |
|
| 0.5014 | 0.96 | 4800 | 0.8118 | |
|
| 0.4701 | 1.0 | 5000 | 0.8114 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.38.2 |
|
- Pytorch 2.2.1+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
