metadata
library_name: peft
tags:
- generated_from_trainer
base_model: bigcode/starcoder2-15b
model-index:
- name: peft-lora-starcoder2-15b-flutter-copilot
results: []
peft-lora-starcoder2-15b-flutter-copilot
This model is a fine-tuned version of bigcode/starcoder2-15b on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.3895
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: 0.0003
- 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 |
|---|---|---|---|
| 0.9837 | 0.05 | 100 | 0.4444 |
| 0.213 | 0.1 | 200 | 0.4486 |
| 0.4039 | 0.15 | 300 | 0.4082 |
| 0.3069 | 0.2 | 400 | 0.4030 |
| 0.3163 | 0.25 | 500 | 0.4044 |
| 0.7574 | 0.3 | 600 | 0.4008 |
| 0.4859 | 0.35 | 700 | 0.3990 |
| 0.5048 | 0.4 | 800 | 0.3984 |
| 0.4226 | 0.45 | 900 | 0.3962 |
| 3.584 | 0.5 | 1000 | 0.3959 |
| 0.624 | 0.55 | 1100 | 0.3958 |
| 0.2663 | 0.6 | 1200 | 0.3954 |
| 0.496 | 0.65 | 1300 | 0.3947 |
| 0.6882 | 0.7 | 1400 | 0.3950 |
| 0.4147 | 0.75 | 1500 | 0.3937 |
| 0.5237 | 0.8 | 1600 | 0.3922 |
| 0.626 | 0.85 | 1700 | 0.3909 |
| 0.6625 | 0.9 | 1800 | 0.3897 |
| 0.6163 | 0.95 | 1900 | 0.3896 |
| 0.5014 | 1.0 | 2000 | 0.3895 |
Framework versions
- PEFT 0.9.1.dev0
- Transformers 4.39.0.dev0
- Pytorch 2.2.1
- Datasets 2.18.0
- Tokenizers 0.15.2