{ "model": "/vol/runs/codex_curriculum/stage1/model", "ref_model": "Qwen/Qwen2.5-0.5B-Instruct", "steps": 1000, "group": 8, "bsz": 1, "lr": 3e-06, "kl_beta": 0.02, "clip_eps": 0.2, "temperature": 1.0, "format_bonus": 0.05, "think_bonus": 0.02, "answer_bonus": 0.0, "numbers_bonus": 0.1, "loss_agg": "seq", "max_new_tokens": 512, "smoke": false, "train_file": "data/train.jsonl", "timestamp": "2026-06-29T08:15:00", "final_mean_reward": 0.2671750111877918, "final_solved_frac": 0.0325, "final_format_rate": 0.95625, "final_within_group_std": 0.10410462675616144, "peak_solved_frac": 1.0, "peak_format_rate": 1.0, "peak_think_rate": 1.0, "peak_answer_rate": 1.0 }