{ "report": { "scenario": "wyrd-scene-fletcher", "cases": 8, "runs": 3, "latency_per_turn": { "e2e_ms": { "median": 179.2, "min": 176.6, "max": 803.0, "n": 24 }, "srv_ms": { "median": 141.1, "min": 139.6, "max": 742.1, "n": 24 } }, "by_decision": { "place": { "evidence": { "n": 8, "correct": 7, "acc": 0.875 }, "blind_null": { "n": 8, "correct": 6, "acc": 0.75 }, "controls": { "n": 3, "correct": 2, "acc": 0.667 }, "non_controls": { "n": 5, "correct": 5, "acc": 1.0 }, "by_label": { "holds": { "evidence": { "n": 6, "correct": 6, "acc": 1.0 }, "blind_null": { "n": 6, "correct": 6, "acc": 1.0 } }, "moved": { "evidence": { "n": 2, "correct": 1, "acc": 0.5 }, "blind_null": { "n": 2, "correct": 0, "acc": 0.0 } } }, "unanimous": { "n": 8, "correct": 7, "acc": 0.875 }, "split": { "n": 0, "correct": 0, "acc": null }, "egregious": 0, "blind_egregious": 0, "misses": [ "f-go-tavern: said holds (p=0.997, agree=1.0) want moved" ] }, "exit": { "evidence": { "n": 8, "correct": 8, "acc": 1.0 }, "blind_null": { "n": 8, "correct": 5, "acc": 0.625 }, "controls": { "n": 3, "correct": 3, "acc": 1.0 }, "non_controls": { "n": 5, "correct": 5, "acc": 1.0 }, "by_label": { "meet_wren": { "evidence": { "n": 2, "correct": 2, "acc": 1.0 }, "blind_null": { "n": 2, "correct": 0, "acc": 0.0 } }, "stay": { "evidence": { "n": 5, "correct": 5, "acc": 1.0 }, "blind_null": { "n": 5, "correct": 5, "acc": 1.0 } }, "tavern": { "evidence": { "n": 1, "correct": 1, "acc": 1.0 }, "blind_null": { "n": 1, "correct": 0, "acc": 0.0 } } }, "unanimous": { "n": 7, "correct": 7, "acc": 1.0 }, "split": { "n": 1, "correct": 1, "acc": 1.0 }, "egregious": 0, "blind_egregious": 0, "misses": [] }, "place2": { "evidence": { "n": 8, "correct": 8, "acc": 1.0 }, "blind_null": { "n": 8, "correct": 6, "acc": 0.75 }, "controls": { "n": 3, "correct": 3, "acc": 1.0 }, "non_controls": { "n": 5, "correct": 5, "acc": 1.0 }, "by_label": { "elsewhere": { "evidence": { "n": 2, "correct": 2, "acc": 1.0 }, "blind_null": { "n": 2, "correct": 0, "acc": 0.0 } }, "same": { "evidence": { "n": 6, "correct": 6, "acc": 1.0 }, "blind_null": { "n": 6, "correct": 6, "acc": 1.0 } } }, "unanimous": { "n": 8, "correct": 8, "acc": 1.0 }, "split": { "n": 0, "correct": 0, "acc": null }, "egregious": 0, "blind_egregious": 0, "misses": [] }, "exit2": { "evidence": { "n": 8, "correct": 5, "acc": 0.625 }, "blind_null": { "n": 8, "correct": 5, "acc": 0.625 }, "controls": { "n": 3, "correct": 1, "acc": 0.333 }, "non_controls": { "n": 5, "correct": 4, "acc": 0.8 }, "by_label": { "meet_wren": { "evidence": { "n": 2, "correct": 0, "acc": 0.0 }, "blind_null": { "n": 2, "correct": 0, "acc": 0.0 } }, "stay": { "evidence": { "n": 5, "correct": 5, "acc": 1.0 }, "blind_null": { "n": 5, "correct": 5, "acc": 1.0 } }, "tavern": { "evidence": { "n": 1, "correct": 0, "acc": 0.0 }, "blind_null": { "n": 1, "correct": 0, "acc": 0.0 } } }, "unanimous": { "n": 8, "correct": 5, "acc": 0.625 }, "split": { "n": 0, "correct": 0, "acc": null }, "egregious": 0, "blind_egregious": 0, "misses": [ "f-go-tavern: said stay (p=0.984, agree=1.0) want tavern", "f-approach-wren: said stay (p=0.993, agree=1.0) want meet_wren", "f-wren-comes: said stay (p=0.996, agree=1.0) want meet_wren" ] } } }, "rows": [ { "case": "f-idle-look", "tag": "control", "decision": "place", "label": "holds", "top": "holds", "p_top": 0.999, "agreement": 1.0, "probs": { "moved": 0.001, "holds": 0.999 }, "blind_top": "holds", "input_tokens": 263, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-idle-look", "tag": "control", "decision": "exit", "label": "stay", "top": "stay", "p_top": 0.908, "agreement": 1.0, "probs": { "stay": 0.908, "meet_wren": 0.071, "tavern": 0.02 }, "blind_top": "stay", "input_tokens": 320, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-idle-look", "tag": "control", "decision": "place2", "label": "same", "top": "same", "p_top": 0.998, "agreement": 1.0, "probs": { "same": 0.998, "elsewhere": 0.002 }, "blind_top": "same", "input_tokens": 276, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-idle-look", "tag": "control", "decision": "exit2", "label": "stay", "top": "stay", "p_top": 1.0, "agreement": 1.0, "probs": { "stay": 1.0, "meet_wren": 0.0, "tavern": 0.0 }, "blind_top": "stay", "input_tokens": 331, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-plan-tavern", "tag": "case", "decision": "place", "label": "holds", "top": "holds", "p_top": 0.999, "agreement": 1.0, "probs": { "moved": 0.001, "holds": 0.999 }, "blind_top": "holds", "input_tokens": 263, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-plan-tavern", "tag": "case", "decision": "exit", "label": "stay", "top": "stay", "p_top": 0.876, "agreement": 1.0, "probs": { "stay": 0.876, "meet_wren": 0.045, "tavern": 0.078 }, "blind_top": "stay", "input_tokens": 320, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-plan-tavern", "tag": "case", "decision": "place2", "label": "same", "top": "same", "p_top": 0.998, "agreement": 1.0, "probs": { "same": 0.998, "elsewhere": 0.002 }, "blind_top": "same", "input_tokens": 276, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-plan-tavern", "tag": "case", "decision": "exit2", "label": "stay", "top": "stay", "p_top": 1.0, "agreement": 1.0, "probs": { "stay": 1.0, "meet_wren": 0.0, "tavern": 0.0 }, "blind_top": "stay", "input_tokens": 331, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-go-tavern", "tag": "control", "decision": "place", "label": "moved", "top": "holds", "p_top": 0.997, "agreement": 1.0, "probs": { "moved": 0.003, "holds": 0.997 }, "blind_top": "holds", "input_tokens": 261, "ok": false, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-go-tavern", "tag": "control", "decision": "exit", "label": "tavern", "top": "tavern", "p_top": 0.924, "agreement": 1.0, "probs": { "stay": 0.062, "meet_wren": 0.014, "tavern": 0.924 }, "blind_top": "stay", "input_tokens": 318, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-go-tavern", "tag": "control", "decision": "place2", "label": "elsewhere", "top": "elsewhere", "p_top": 0.977, "agreement": 1.0, "probs": { "same": 0.023, "elsewhere": 0.977 }, "blind_top": "same", "input_tokens": 274, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-go-tavern", "tag": "control", "decision": "exit2", "label": "tavern", "top": "stay", "p_top": 0.984, "agreement": 1.0, "probs": { "stay": 0.984, "meet_wren": 0.004, "tavern": 0.012 }, "blind_top": "stay", "input_tokens": 329, "ok": false, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-approach-wren", "tag": "control", "decision": "place", "label": "holds", "top": "holds", "p_top": 0.999, "agreement": 1.0, "probs": { "moved": 0.001, "holds": 0.999 }, "blind_top": "holds", "input_tokens": 263, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-approach-wren", "tag": "control", "decision": "exit", "label": "meet_wren", "top": "meet_wren", "p_top": 0.988, "agreement": 1.0, "probs": { "stay": 0.01, "meet_wren": 0.988, "tavern": 0.002 }, "blind_top": "stay", "input_tokens": 320, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-approach-wren", "tag": "control", "decision": "place2", "label": "same", "top": "same", "p_top": 0.995, "agreement": 1.0, "probs": { "same": 0.995, "elsewhere": 0.005 }, "blind_top": "same", "input_tokens": 276, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-approach-wren", "tag": "control", "decision": "exit2", "label": "meet_wren", "top": "stay", "p_top": 0.993, "agreement": 1.0, "probs": { "stay": 0.993, "meet_wren": 0.007, "tavern": 0.0 }, "blind_top": "stay", "input_tokens": 331, "ok": false, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-memory", "tag": "case", "decision": "place", "label": "holds", "top": "holds", "p_top": 1.0, "agreement": 1.0, "probs": { "moved": 0.0, "holds": 1.0 }, "blind_top": "holds", "input_tokens": 265, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-memory", "tag": "case", "decision": "exit", "label": "stay", "top": "stay", "p_top": 0.862, "agreement": 1.0, "probs": { "stay": 0.862, "meet_wren": 0.117, "tavern": 0.021 }, "blind_top": "stay", "input_tokens": 322, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-memory", "tag": "case", "decision": "place2", "label": "same", "top": "same", "p_top": 0.999, "agreement": 1.0, "probs": { "same": 0.999, "elsewhere": 0.001 }, "blind_top": "same", "input_tokens": 278, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-memory", "tag": "case", "decision": "exit2", "label": "stay", "top": "stay", "p_top": 1.0, "agreement": 1.0, "probs": { "stay": 1.0, "meet_wren": 0.0, "tavern": 0.0 }, "blind_top": "stay", "input_tokens": 333, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-leave-town", "tag": "case", "decision": "place", "label": "moved", "top": "moved", "p_top": 0.766, "agreement": 1.0, "probs": { "moved": 0.766, "holds": 0.234 }, "blind_top": "holds", "input_tokens": 259, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-leave-town", "tag": "case", "decision": "exit", "label": "stay", "top": "stay", "p_top": 0.667, "agreement": 0.6666666666666666, "probs": { "stay": 0.667, "meet_wren": 0.29, "tavern": 0.043 }, "blind_top": "stay", "input_tokens": 316, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-leave-town", "tag": "case", "decision": "place2", "label": "elsewhere", "top": "elsewhere", "p_top": 0.999, "agreement": 1.0, "probs": { "same": 0.001, "elsewhere": 0.999 }, "blind_top": "same", "input_tokens": 272, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-leave-town", "tag": "case", "decision": "exit2", "label": "stay", "top": "stay", "p_top": 0.989, "agreement": 1.0, "probs": { "stay": 0.989, "meet_wren": 0.009, "tavern": 0.002 }, "blind_top": "stay", "input_tokens": 327, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-nightfall", "tag": "case", "decision": "place", "label": "holds", "top": "holds", "p_top": 1.0, "agreement": 1.0, "probs": { "moved": 0.0, "holds": 1.0 }, "blind_top": "holds", "input_tokens": 248, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-nightfall", "tag": "case", "decision": "exit", "label": "stay", "top": "stay", "p_top": 0.988, "agreement": 1.0, "probs": { "stay": 0.988, "meet_wren": 0.01, "tavern": 0.003 }, "blind_top": "stay", "input_tokens": 305, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-nightfall", "tag": "case", "decision": "place2", "label": "same", "top": "same", "p_top": 1.0, "agreement": 1.0, "probs": { "same": 1.0, "elsewhere": 0.0 }, "blind_top": "same", "input_tokens": 261, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-nightfall", "tag": "case", "decision": "exit2", "label": "stay", "top": "stay", "p_top": 1.0, "agreement": 1.0, "probs": { "stay": 1.0, "meet_wren": 0.0, "tavern": 0.0 }, "blind_top": "stay", "input_tokens": 316, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-wren-comes", "tag": "case", "decision": "place", "label": "holds", "top": "holds", "p_top": 0.999, "agreement": 1.0, "probs": { "moved": 0.001, "holds": 0.999 }, "blind_top": "holds", "input_tokens": 268, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-wren-comes", "tag": "case", "decision": "exit", "label": "meet_wren", "top": "meet_wren", "p_top": 0.68, "agreement": 1.0, "probs": { "stay": 0.296, "meet_wren": 0.68, "tavern": 0.024 }, "blind_top": "stay", "input_tokens": 325, "ok": true, "blind_ok": false, "egregious": false, "blind_egregious": false }, { "case": "f-wren-comes", "tag": "case", "decision": "place2", "label": "same", "top": "same", "p_top": 0.999, "agreement": 1.0, "probs": { "same": 0.999, "elsewhere": 0.001 }, "blind_top": "same", "input_tokens": 281, "ok": true, "blind_ok": true, "egregious": false, "blind_egregious": false }, { "case": "f-wren-comes", "tag": "case", "decision": "exit2", "label": "meet_wren", "top": "stay", "p_top": 0.996, "agreement": 1.0, "probs": { "stay": 0.996, "meet_wren": 0.004, "tavern": 0.001 }, "blind_top": "stay", "input_tokens": 336, "ok": false, "blind_ok": false, "egregious": false, "blind_egregious": false } ] }