---
license: mit
base_model: Qwen/Qwen2.5-7B-Instruct
tags:
  - debugging
  - tool-use
  - multi-turn
  - reinforcement-learning
  - grpo
datasets:
  - custom
language:
  - en
pipeline_tag: text-generation
---

# DSL Debug 7B — RL-Only Step 30

Qwen2.5-7B-Instruct trained with GRPO (Group Relative Policy Optimization) directly from the base model, no SFT warmup.

**Blog post:** [Multi-Turn RL for Code Debugging](https://andrewlngdn.github.io/dsl_debugger/)
**Code + environment:** [github.com/AndrewLngdn/dsl-debug](https://github.com/AndrewLngdn/dsl-debug)

## Training

- **Method**: GRPO with multi-turn tool use (verl 0.7 + sglang 0.5.6)
- **Base model**: Qwen2.5-7B-Instruct (no SFT warmup)
- **Steps**: 30 (batch size 512, 8 rollouts per prompt)
- **LR**: 1e-5 cosine
- **Reward**: Binary (1.0 if submitted code matches expected output, 0.0 otherwise)
- **Hardware**: 2x A100-SXM4-80GB

## Results (held-out test, one-shot)

| Split | Base Model | This Model |
|-------|:---:|:---:|
| Standard (481) | 50.5% | **78.8%** |
| Nonlocal (200) | 12.0% | **54.0%** |
| Intent-Mismatch (177) | 0.6% | **14.7%** |

## Alignment Tax

| Benchmark | Base | This Model |
|-----------|:---:|:---:|
| MMLU (5-shot) | 74.6% | 74.7% |
| GSM8K (8-shot) | 84.9% | 84.4% |
| HumanEval (0-shot) | 65.9% | 59.1% |

## Usage

```python
from huggingface_hub import snapshot_download
snapshot_download("andrewlngdn/dsl-debug-7b-rl-only-step30",
    local_dir="/workspace/models/rl_only_step30")
```

## Related Models

| Model | Repo |
|-------|------|
| **SFT then RL step 35 (best)** | [andrewlngdn/dsl-debug-7b-sft-rl](https://huggingface.co/andrewlngdn/dsl-debug-7b-sft-rl) |
| SFT step 100 | [andrewlngdn/dsl-debug-7b-sft-step100](https://huggingface.co/andrewlngdn/dsl-debug-7b-sft-step100) |