{ "method": "SimPO-avg", "loss_type": "simpo", "seed": 42, "backbone": "Qwen/Qwen3-8B", "base_model_id": "Qwen/Qwen3-8B", "simpo_beta": 2.5, "simpo_gamma": 1.0, "status": "completed", "returncode": 0, "completed_at": "2026-07-08T08:48:44Z", "dataset": "/data/mnpo/revision_qwen3_8b/full_iter1/precomputed/avg_oracle", "output_dir": "/data/mnpo/revision_qwen3_8b/full_iter1/train/simpo_avg_b2p5_g1p0_s42_h200", "config": "/data/mnpo/revision_qwen3_8b/full_iter1/train/simpo_avg_b2p5_g1p0_s42_h200/config.yaml", "train_metrics_exists": true, "trainer_state_exists": true }