metadata
license: apache-2.0
base_model: Qwen/Qwen3-0.6B
datasets:
- nvidia/OpenCodeReasoning
tags:
- code-generation
- reasoning
P4o1o/Qwen3_0.6-NvidiaOpenCodeReasoning
Fine-tuned version of Qwen/Qwen3-0.6B on NVIDIA's OpenCodeReasoning dataset
Training Hyperparameters
- Batch size: 16
- Learning rate: 1e-05
- Epochs: 3
- Max length: 2048