metadata
base_model: bigcode/starcoderbase-1b
library_name: peft
license: bigcode-openrail-m
tags:
- generated_from_trainer
model-index:
- name: peft-starcoder-lora-a100
results: []
peft-starcoder-lora-a100
This model is a fine-tuned version of bigcode/starcoderbase-1b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.5624
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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 30
- training_steps: 2000
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.2574 | 0.05 | 100 | 0.9670 |
| 0.1994 | 0.1 | 200 | 1.0338 |
| 0.1515 | 0.15 | 300 | 1.1043 |
| 0.1391 | 0.2 | 400 | 1.1603 |
| 0.1373 | 0.25 | 500 | 1.2171 |
| 0.1007 | 0.3 | 600 | 1.2480 |
| 0.0987 | 0.35 | 700 | 1.2648 |
| 0.0811 | 0.4 | 800 | 1.2875 |
| 0.0743 | 0.45 | 900 | 1.3243 |
| 0.0694 | 0.5 | 1000 | 1.3702 |
| 0.0565 | 0.55 | 1100 | 1.3839 |
| 0.0631 | 0.6 | 1200 | 1.4482 |
| 0.0511 | 0.65 | 1300 | 1.4734 |
| 0.0436 | 0.7 | 1400 | 1.5073 |
| 0.0415 | 0.75 | 1500 | 1.5295 |
| 0.0394 | 0.8 | 1600 | 1.5387 |
| 0.0374 | 0.85 | 1700 | 1.5535 |
| 0.0383 | 0.9 | 1800 | 1.5564 |
| 0.0364 | 0.95 | 1900 | 1.5632 |
| 0.041 | 1.0 | 2000 | 1.5624 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1