metadata
license: bigcode-openrail-m
library_name: peft
tags:
- generated_from_trainer
base_model: bigcode/starcoder
model-index:
- name: Starcoder-StaproCoder
results: []
Starcoder-StaproCoder
This model is a fine-tuned version of bigcode/starcoder on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2404
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 |
|---|---|---|---|
| 0.3529 | 0.05 | 100 | 0.3017 |
| 0.3508 | 0.1 | 200 | 0.3189 |
| 0.3033 | 0.15 | 300 | 0.2719 |
| 0.2647 | 0.2 | 400 | 0.2645 |
| 0.2757 | 0.25 | 500 | 0.2583 |
| 0.1933 | 0.3 | 600 | 0.2514 |
| 0.2572 | 0.35 | 700 | 0.2471 |
| 0.2468 | 0.4 | 800 | 0.2496 |
| 0.2772 | 0.45 | 900 | 0.2444 |
| 0.2317 | 0.5 | 1000 | 0.2500 |
| 0.1994 | 0.55 | 1100 | 0.2452 |
| 0.2172 | 0.6 | 1200 | 0.2383 |
| 0.2098 | 0.65 | 1300 | 0.2409 |
| 0.2156 | 0.7 | 1400 | 0.2383 |
| 0.1733 | 0.75 | 1500 | 0.2417 |
| 0.1801 | 0.8 | 1600 | 0.2400 |
| 0.1295 | 0.85 | 1700 | 0.2417 |
| 0.1331 | 0.9 | 1800 | 0.2399 |
| 0.1997 | 0.95 | 1900 | 0.2406 |
| 0.1827 | 1.0 | 2000 | 0.2404 |
Framework versions
- PEFT 0.9.0
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2