peft-lora-starcoder15B-v2-personal-copilot-A100-40GB-colab
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: 0.1653
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_steps: 30
- training_steps: 2000
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.5026 | 0.05 | 100 | 0.3368 |
0.4842 | 0.1 | 200 | 0.2902 |
0.4734 | 0.15 | 300 | 0.2565 |
0.4208 | 0.2 | 400 | 0.2278 |
0.4774 | 0.25 | 500 | 0.2273 |
0.7966 | 0.3 | 600 | 0.2202 |
0.3598 | 0.35 | 700 | 0.1997 |
0.4743 | 0.4 | 800 | 0.2000 |
0.355 | 0.45 | 900 | 0.1891 |
0.3507 | 0.5 | 1000 | 0.1735 |
0.5402 | 0.55 | 1100 | 0.1733 |
0.4928 | 0.6 | 1200 | 0.1721 |
0.5689 | 0.65 | 1300 | 0.1668 |
0.4705 | 0.7 | 1400 | 0.1672 |
0.6738 | 0.75 | 1500 | 0.1672 |
0.3904 | 0.8 | 1600 | 0.1666 |
0.4451 | 0.85 | 1700 | 0.1668 |
0.418 | 0.9 | 1800 | 0.1652 |
0.418 | 0.95 | 1900 | 0.1653 |
0.3868 | 1.0 | 2000 | 0.1653 |
Framework versions
- PEFT 0.7.1
- Transformers 4.36.2
- Pytorch 2.1.2+cu121
- Datasets 2.16.1
- Tokenizers 0.15.0
- Downloads last month
- 3
Model tree for ZiHDeng/peft-lora-starcoder15B-v2-personal-copilot-A100-40GB-colab
Base model
bigcode/starcoderbase-1b