{
  "tag": "full",
  "n_tasks": 100,
  "k": 8,
  "budgets": [
    "no_think",
    "256",
    "512",
    "1024",
    "2048",
    "unbudgeted"
  ],
  "controls": false,
  "model": "Qwen/Qwen3.5-4B",
  "gen_secs": 11139.2,
  "conditions": {
    "shuffle_512": {
      "n_tasks": 100,
      "greedy_pass@1": 0.8,
      "sampled_pass@1": 0.7925,
      "pass@8_oracle": 0.93,
      "visible_selector@1": 0.89,
      "oracle_minus_deployable": 0.04,
      "mean_think_tokens": 411.4,
      "mean_total_tokens": 491.2,
      "forced_close_frac": 0.429
    },
    "shuffle_2048": {
      "n_tasks": 100,
      "greedy_pass@1": 0.86,
      "sampled_pass@1": 0.8662,
      "pass@8_oracle": 0.94,
      "visible_selector@1": 0.92,
      "oracle_minus_deployable": 0.02,
      "mean_think_tokens": 607.6,
      "mean_total_tokens": 681.1,
      "forced_close_frac": 0.065
    }
  }
}