peft-lora-starcoder1B-Instruction-ny8-0204
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: 0.2566
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
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
0.1712 | 0.05 | 100 | 0.1588 |
0.1403 | 0.1 | 200 | 0.1454 |
0.1231 | 0.15 | 300 | 0.1525 |
0.1094 | 0.2 | 400 | 0.1705 |
0.1013 | 0.25 | 500 | 0.1767 |
0.092 | 0.3 | 600 | 0.1910 |
0.0877 | 0.35 | 700 | 0.1999 |
0.0813 | 0.4 | 800 | 0.2090 |
0.0802 | 0.45 | 900 | 0.2203 |
0.0776 | 0.5 | 1000 | 0.2261 |
0.073 | 0.55 | 1100 | 0.2387 |
0.0708 | 0.6 | 1200 | 0.2387 |
0.0698 | 0.65 | 1300 | 0.2459 |
0.068 | 0.7 | 1400 | 0.2477 |
0.0677 | 0.75 | 1500 | 0.2507 |
0.0645 | 0.8 | 1600 | 0.2515 |
0.0648 | 0.85 | 1700 | 0.2584 |
0.0645 | 0.9 | 1800 | 0.2562 |
0.0643 | 0.95 | 1900 | 0.2559 |
0.0642 | 1.0 | 2000 | 0.2566 |
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
- 1
Model tree for ZiHDeng/peft-lora-starcoder1B-Instruction-ny8-0204
Base model
bigcode/starcoderbase-1b