---
license: apache-2.0
base_model: hamishivi/sft_qwen3_8b_our_sft
library_name: transformers
tags:
  - open-instruct
  - grpo
  - qwen3
  - swerl
---

# SWERL Qwen3 8B TMAX 15k GRPO

Step 500 checkpoint from the `hamishivi/swerl-tmax-15k` GRPO run.

- Base model: `hamishivi/sft_qwen3_8b_our_sft`
- Run ID: `swerl_qwen3_8b_our_sft_tmax_15k_env_prewarm_offline__seed42__20260516_091834`
- Step: `500`

This checkpoint is intended for internal evaluation and continuation experiments.