This is the model that is used to get the paper results for the MPCxR1 Qwen2.5 3B SFT GRPO model.
This model was evaluated on the 19.04.25. and trained on the 18.04.25 at 13:29:26.
The base model used for this training was "nibauman/race_llm_Qwen_3B_sft"
here
This is the wandb training:
https://wandb.ai/CoRL-heist-2025/mpc_grpo/runs/osa45as5?nw=nwusernibaumaneth
