metadata
library_name: peft
license: apache-2.0
base_model: unsloth/qwen2.5-coder-32b-instruct-bnb-4bit
tags:
- trl
- sft
- unsloth
- generated_from_trainer
model-index:
- name: Qwen-coder-32B-Instruct_playpen_SFT_DFINAL_1.7K-steps
results: []
Qwen-coder-32B-Instruct_playpen_SFT_DFINAL_1.7K-steps
This model is a fine-tuned version of unsloth/qwen2.5-coder-32b-instruct-bnb-4bit on the None dataset. It achieves the following results on the evaluation set:
- Loss: 2.6435
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: 0.0002
- train_batch_size: 4
- eval_batch_size: 4
- seed: 7331
- optimizer: Use adamw_8bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.03
- lr_scheduler_warmup_steps: 5
- training_steps: 1700
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.0006 | 0.0565 | 100 | 2.3866 |
| 0.001 | 0.1130 | 200 | 2.3386 |
| 0.0009 | 0.1695 | 300 | 2.4257 |
| 0.0007 | 0.2260 | 400 | 2.4474 |
| 0.001 | 0.2825 | 500 | 2.5487 |
| 0.0008 | 0.3390 | 600 | 2.5405 |
| 0.0009 | 0.3955 | 700 | 2.5610 |
| 0.0017 | 0.4520 | 800 | 2.5255 |
| 0.0008 | 0.5085 | 900 | 2.5441 |
| 0.0007 | 0.5650 | 1000 | 2.6643 |
| 0.0006 | 0.6215 | 1100 | 2.7327 |
| 0.0006 | 0.6780 | 1200 | 2.7778 |
| 0.0008 | 0.7345 | 1300 | 2.6578 |
| 0.0006 | 0.7910 | 1400 | 2.6902 |
| 0.001 | 0.8475 | 1500 | 2.6243 |
| 0.0009 | 0.9040 | 1600 | 2.6423 |
| 0.0008 | 0.9605 | 1700 | 2.6435 |
Framework versions
- PEFT 0.14.0
- Transformers 4.47.1
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.21.0