metadata
license: bigscience-openrail-m
library_name: peft
tags:
- generated_from_trainer
base_model: WizardLM/WizardCoder-15B-V1.0
model-index:
- name: Wizardcoder13B-StaproCoder
results: []
Wizardcoder13B-StaproCoder
This model is a fine-tuned version of WizardLM/WizardCoder-15B-V1.0 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2314
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: 0.0005
- 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_ratio: 0.1
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.2997 | 0.05 | 100 | 0.3043 |
| 0.3811 | 0.1 | 200 | 0.2913 |
| 0.242 | 0.15 | 300 | 0.2843 |
| 0.2858 | 0.2 | 400 | 0.2756 |
| 0.1813 | 0.25 | 500 | 0.2758 |
| 0.2717 | 0.3 | 600 | 0.2656 |
| 0.2431 | 0.35 | 700 | 0.2660 |
| 0.3395 | 0.4 | 800 | 0.4781 |
| 0.3274 | 0.45 | 900 | 0.2493 |
| 0.1886 | 0.5 | 1000 | 0.2451 |
| 0.1963 | 0.55 | 1100 | 0.2422 |
| 0.2548 | 0.6 | 1200 | 0.2401 |
| 0.3505 | 0.65 | 1300 | 0.2363 |
| 0.2788 | 0.7 | 1400 | 0.2344 |
| 0.186 | 0.75 | 1500 | 0.2323 |
| 0.1847 | 0.8 | 1600 | 0.2325 |
| 0.2109 | 0.85 | 1700 | 0.2313 |
| 0.2144 | 0.9 | 1800 | 0.2316 |
| 0.2038 | 0.95 | 1900 | 0.2316 |
| 0.2071 | 1.0 | 2000 | 0.2314 |
Framework versions
- PEFT 0.9.0
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2