{
  "passed": true,
  "cases": 24,
  "predictions": 264,
  "chunks": 424,
  "exact_repeat_chunks": 56,
  "all_mp4_decoded": 528,
  "independent_proprio_feedback_checks": 160,
  "initial_inputs_matched": 264,
  "summary": [
    {
      "cohort": "episode_start",
      "chunks": 1,
      "name": "alternative",
      "pairs": 32,
      "text_xyz_mm_median": 9.094492170620107,
      "seed_xyz_mm_median": 40.047061730124426,
      "text_rgb_mae_median": 5.633874893188477,
      "seed_rgb_mae_median": 19.496986389160156,
      "text_active_rmse_median": 0.03906535044359605,
      "seed_active_rmse_median": 0.22504594086677124,
      "code_changed_fraction": 0.9375,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "episode_start",
      "chunks": 1,
      "name": "stage_only",
      "pairs": 32,
      "text_xyz_mm_median": 16.10552397787837,
      "seed_xyz_mm_median": 40.047061730124426,
      "text_rgb_mae_median": 12.187390804290771,
      "seed_rgb_mae_median": 19.496986389160156,
      "text_active_rmse_median": 0.09622692941937216,
      "seed_active_rmse_median": 0.22504594086677124,
      "code_changed_fraction": 1.0,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "episode_start",
      "chunks": 1,
      "name": "task_only",
      "pairs": 32,
      "text_xyz_mm_median": 17.130510841061913,
      "seed_xyz_mm_median": 40.047061730124426,
      "text_rgb_mae_median": 11.377955436706543,
      "seed_rgb_mae_median": 19.496986389160156,
      "text_active_rmse_median": 0.11356648855588862,
      "seed_active_rmse_median": 0.22504594086677124,
      "code_changed_fraction": 1.0,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "episode_start",
      "chunks": 1,
      "name": "hold",
      "pairs": 32,
      "text_xyz_mm_median": 17.187276039059736,
      "seed_xyz_mm_median": 40.047061730124426,
      "text_rgb_mae_median": 12.086161136627197,
      "seed_rgb_mae_median": 19.496986389160156,
      "text_active_rmse_median": 0.0994079349112402,
      "seed_active_rmse_median": 0.22504594086677124,
      "code_changed_fraction": 1.0,
      "hold_motion_ratio_median": 1.0176036886574895,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "episode_start",
      "chunks": 3,
      "name": "alternative",
      "pairs": 32,
      "text_xyz_mm_median": 33.146518421576985,
      "seed_xyz_mm_median": 85.99843297438838,
      "text_rgb_mae_median": 9.068046013514202,
      "seed_rgb_mae_median": 23.90681012471517,
      "text_active_rmse_median": 0.1278887046190364,
      "seed_active_rmse_median": 0.285796042460658,
      "code_changed_fraction": 0.9375,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "stage",
      "chunks": 1,
      "name": "alternative",
      "pairs": 16,
      "text_xyz_mm_median": 11.456958987154621,
      "seed_xyz_mm_median": 34.84474583515421,
      "text_rgb_mae_median": 6.805502653121948,
      "seed_rgb_mae_median": 18.297040939331055,
      "text_active_rmse_median": 0.046401200033182405,
      "seed_active_rmse_median": 0.1557747458281781,
      "code_changed_fraction": 1.0,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "stage",
      "chunks": 1,
      "name": "stage_only",
      "pairs": 16,
      "text_xyz_mm_median": 26.894861071396758,
      "seed_xyz_mm_median": 34.84474583515421,
      "text_rgb_mae_median": 12.624500751495361,
      "seed_rgb_mae_median": 18.297040939331055,
      "text_active_rmse_median": 0.11929243743537757,
      "seed_active_rmse_median": 0.1557747458281781,
      "code_changed_fraction": 1.0,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "stage",
      "chunks": 1,
      "name": "task_only",
      "pairs": 16,
      "text_xyz_mm_median": 23.610193502290244,
      "seed_xyz_mm_median": 34.84474583515421,
      "text_rgb_mae_median": 14.927546501159668,
      "seed_rgb_mae_median": 18.297040939331055,
      "text_active_rmse_median": 0.13190888205339107,
      "seed_active_rmse_median": 0.1557747458281781,
      "code_changed_fraction": 1.0,
      "seed_code_changed_fraction": 0.0
    },
    {
      "cohort": "stage",
      "chunks": 1,
      "name": "hold",
      "pairs": 16,
      "text_xyz_mm_median": 15.221352697275623,
      "seed_xyz_mm_median": 34.84474583515421,
      "text_rgb_mae_median": 12.024145603179932,
      "seed_rgb_mae_median": 18.297040939331055,
      "text_active_rmse_median": 0.06641824835528731,
      "seed_active_rmse_median": 0.1557747458281781,
      "code_changed_fraction": 1.0,
      "hold_motion_ratio_median": 0.9805140013815543,
      "seed_code_changed_fraction": 0.0
    }
  ],
  "feedback_clipped_predictions": 0,
  "closed_loop": false,
  "semantic_success": false,
  "cross_worker_embedding_hashes_match": true,
  "embedding_prompts": 41
}