Download results/eval.json from psiloceyeben/oracle7-deepspace: direct link, hf CLI and curl.
- Browser
- Download file 14.9 kB
-
https://huggingface.co/psiloceyeben/oracle7-deepspace/resolve/main/results/eval.json
- Command line
-
hf download hf://psiloceyeben/oracle7-deepspace/results/eval.json
-
curl -L -o eval.json https://huggingface.co/psiloceyeben/oracle7-deepspace/resolve/main/results/eval.json
14.9 kB
| { | |
| "oracle7-deepspace": { | |
| "total": { | |
| "correct": 101, | |
| "harmful": 2, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 108, | |
| "tick_ms_mean": 0.054, | |
| "latency_min_median": 3, | |
| "within_20_min": 53, | |
| "twin_runs": 418 | |
| }, | |
| "per_fault": { | |
| "nominal": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_on": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_off": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "wheel_friction": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "sensor_bias": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "star_tracker_dropout": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "solar_degradation": { | |
| "correct": 7, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "battery_short": { | |
| "correct": 10, | |
| "harmful": 2, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "seu_spike": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| } | |
| } | |
| }, | |
| "baseline: limit->safe mode": { | |
| "total": { | |
| "correct": 36, | |
| "harmful": 12, | |
| "safe_mode": 84, | |
| "damage": 8, | |
| "n": 108, | |
| "tick_ms_mean": 0.06, | |
| "latency_min_median": 3, | |
| "within_20_min": 12, | |
| "twin_runs": 0 | |
| }, | |
| "per_fault": { | |
| "nominal": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_on": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_off": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 5, | |
| "n": 12 | |
| }, | |
| "wheel_friction": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "sensor_bias": { | |
| "correct": 0, | |
| "harmful": 12, | |
| "safe_mode": 12, | |
| "damage": 3, | |
| "n": 12 | |
| }, | |
| "star_tracker_dropout": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "solar_degradation": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "battery_short": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "seu_spike": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| } | |
| } | |
| }, | |
| "no action": { | |
| "total": { | |
| "correct": 24, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 5, | |
| "n": 108, | |
| "tick_ms_mean": 0.016, | |
| "latency_min_median": null, | |
| "within_20_min": 0, | |
| "twin_runs": 0 | |
| }, | |
| "per_fault": { | |
| "nominal": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_on": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_off": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 5, | |
| "n": 12 | |
| }, | |
| "wheel_friction": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "sensor_bias": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "star_tracker_dropout": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "solar_degradation": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "battery_short": { | |
| "correct": 0, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "seu_spike": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| } | |
| } | |
| }, | |
| "random admissible response": { | |
| "total": { | |
| "correct": 48, | |
| "harmful": 9, | |
| "safe_mode": 26, | |
| "damage": 5, | |
| "n": 108, | |
| "tick_ms_mean": 0.038, | |
| "latency_min_median": 3, | |
| "within_20_min": 20, | |
| "twin_runs": 0 | |
| }, | |
| "per_fault": { | |
| "nominal": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_on": { | |
| "correct": 2, | |
| "harmful": 0, | |
| "safe_mode": 4, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_off": { | |
| "correct": 2, | |
| "harmful": 0, | |
| "safe_mode": 4, | |
| "damage": 4, | |
| "n": 12 | |
| }, | |
| "wheel_friction": { | |
| "correct": 6, | |
| "harmful": 2, | |
| "safe_mode": 3, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "sensor_bias": { | |
| "correct": 6, | |
| "harmful": 5, | |
| "safe_mode": 3, | |
| "damage": 1, | |
| "n": 12 | |
| }, | |
| "star_tracker_dropout": { | |
| "correct": 2, | |
| "harmful": 0, | |
| "safe_mode": 4, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "solar_degradation": { | |
| "correct": 2, | |
| "harmful": 0, | |
| "safe_mode": 4, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "battery_short": { | |
| "correct": 4, | |
| "harmful": 2, | |
| "safe_mode": 4, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "seu_spike": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| } | |
| } | |
| }, | |
| "gate_log": [ | |
| { | |
| "rule": "MUTANT-bias->heater", | |
| "verdict": "REJECT", | |
| "reasons": [ | |
| "loses correct responses 35 -> 31", | |
| "adds harmful responses 1 -> 5" | |
| ], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 152, | |
| "ms_total": 540.3, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 31, | |
| "harmful": 5, | |
| "twin_runs": 136, | |
| "ms_total": 628.0, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "MUTANT-wheel->safe", | |
| "verdict": "REJECT", | |
| "reasons": [ | |
| "loses correct responses 35 -> 31" | |
| ], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 152, | |
| "ms_total": 539.0, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 31, | |
| "harmful": 1, | |
| "twin_runs": 138, | |
| "ms_total": 532.4, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "MUTANT-solar->safe", | |
| "verdict": "REJECT", | |
| "reasons": [ | |
| "loses correct responses 35 -> 31" | |
| ], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 152, | |
| "ms_total": 547.1, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 31, | |
| "harmful": 1, | |
| "twin_runs": 130, | |
| "ms_total": 547.8, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-heater_on_uncommanded-switch_heater_backup", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 152, | |
| "ms_total": 532.3, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 140, | |
| "ms_total": 527.1, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-heater_off_uncommanded-switch_heater_backup", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 140, | |
| "ms_total": 526.2, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 124, | |
| "ms_total": 556.8, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-wheel_overcurrent-switch_to_spare_wheel", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 124, | |
| "ms_total": 545.4, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 108, | |
| "ms_total": 514.1, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-sensor_disagree-mark_sensor_invalid", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 108, | |
| "ms_total": 515.2, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 92, | |
| "ms_total": 554.7, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-attitude_knowledge_lost-switch_attitude_sensor", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 92, | |
| "ms_total": 494.5, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 80, | |
| "ms_total": 492.5, | |
| "n": 36 | |
| } | |
| }, | |
| { | |
| "rule": "R-solar_low-shed_loads", | |
| "verdict": "ADMIT", | |
| "reasons": [], | |
| "baseline": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 80, | |
| "ms_total": 493.2, | |
| "n": 36 | |
| }, | |
| "candidate": { | |
| "correct": 35, | |
| "harmful": 1, | |
| "twin_runs": 56, | |
| "ms_total": 502.5, | |
| "n": 36 | |
| } | |
| } | |
| ], | |
| "deepspace + learned rules": { | |
| "total": { | |
| "correct": 106, | |
| "harmful": 2, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 108, | |
| "tick_ms_mean": 0.044, | |
| "latency_min_median": 3, | |
| "within_20_min": 58, | |
| "twin_runs": 140 | |
| }, | |
| "per_fault": { | |
| "nominal": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_on": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "heater_stuck_off": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "wheel_friction": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "sensor_bias": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "star_tracker_dropout": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "solar_degradation": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "battery_short": { | |
| "correct": 10, | |
| "harmful": 2, | |
| "safe_mode": 12, | |
| "damage": 0, | |
| "n": 12 | |
| }, | |
| "seu_spike": { | |
| "correct": 12, | |
| "harmful": 0, | |
| "safe_mode": 0, | |
| "damage": 0, | |
| "n": 12 | |
| } | |
| } | |
| }, | |
| "self_report": { | |
| "real": 15, | |
| "n": 15, | |
| "fakes": { | |
| "echo": 0, | |
| "canned 'all nominal'": 0, | |
| "always decline": 4 | |
| }, | |
| "items": [ | |
| { | |
| "q": "status?", | |
| "a": "Mode: science on. Battery 99%, net power +2.6 A. Temperatures: battery 12.6 C, avionics 12.1 C, tank 16.2 C. Pointing error 0.18 deg. [tm t=320] No open anomaly.", | |
| "must": [ | |
| "Mode:", | |
| "Battery", | |
| "[tm t=" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what is wrong?", | |
| "a": "Recorded faults: at t=112 the battery kept heating while its heater was commanded off, so I switched the battery heater to its backup circuit [receipt #0].", | |
| "must": [ | |
| "heater", | |
| "backup circuit", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "why did you switch the heater?", | |
| "a": "At t=112 I switched the battery heater to its backup circuit because the battery kept heating while its heater was commanded off. I chose it by simulating each option on my onboard model (worst-case scores, lower is better: switch heater backup scored 2.0, collect evidence scored 69.0, safe mode scored 109.0). [receipt #0]", | |
| "must": [ | |
| "because", | |
| "kept heating", | |
| "worst-case", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what is the battery temperature?", | |
| "a": "The battery is at 12.6 C (sensor A 12.3, sensor B 12.8). [tm t=320]", | |
| "must": [ | |
| "battery is at", | |
| "sensor A", | |
| "[tm t=" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "why did you enter safe mode?", | |
| "a": "I have no onboard record of taking that action.", | |
| "must": [ | |
| "no onboard record" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "who won the 1998 world cup?", | |
| "a": "I have no onboard record that answers that.", | |
| "must": [ | |
| "no onboard record" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "any faults?", | |
| "a": "Recorded faults: at t=53 the two avionics sensors disagreed and sensor B moved away from the zone's history, so I marked avionics sensor B invalid [receipt #1].", | |
| "must": [ | |
| "sensor", | |
| "invalid", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "why did you mark the sensor invalid?", | |
| "a": "At t=53 I marked avionics sensor B invalid because the two avionics sensors disagreed and sensor B moved away from the zone's history. I chose it by simulating each option on my onboard model (worst-case scores, lower is better: mark sensor invalid scored 1.0, collect evidence scored 10.0, switch heater backup scored 12.0). [receipt #1]", | |
| "must": [ | |
| "disagreed", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "how confident are you?", | |
| "a": "My last scored decision (t=53, mark sensor invalid) beat the next option by 9.0 points of worst-case cost. [receipt #1]", | |
| "must": [ | |
| "beat the next option", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what is the temperature on europa?", | |
| "a": "I have no onboard record that answers that.", | |
| "must": [ | |
| "no onboard record" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what is wrong?", | |
| "a": "No confirmed fault. At t=41 I saw a single-tick spike on battery sensor A and waited; it did not persist, so I took no action. [receipt #0]", | |
| "must": [ | |
| "No confirmed fault", | |
| "did not persist" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "why did you shed the science load?", | |
| "a": "I have no onboard record of taking that action.", | |
| "must": [ | |
| "no onboard record" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "why did you switch the wheel?", | |
| "a": "At t=64 I switched reaction wheel 1 to the spare because reaction wheel 1 drew more than 1 A. I chose it by simulating each option on my onboard model (worst-case scores, lower is better: switch to spare wheel scored 2.0, collect evidence scored 55.0, shed loads scored 63.0). [receipt #0]", | |
| "must": [ | |
| "reaction wheel", | |
| "more than 1 A", | |
| "[receipt #" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what is the pointing?", | |
| "a": "Pointing error is 0.12 deg using the star tracker; wheels [0, 3, 2] active. [tm t=320]", | |
| "must": [ | |
| "Pointing error", | |
| "[tm t=" | |
| ], | |
| "ok": true | |
| }, | |
| { | |
| "q": "what did you learn?", | |
| "a": "I admitted 6 rule(s) and rejected 3. R-heater_on_uncommanded-switch_heater_backup: validation suite correct 35->35, harmful 1->1, twin runs 152->140. R-heater_off_uncommanded-switch_heater_backup: validation suite correct 35->35, harmful 1->1, twin runs 140->124. R-wheel_overcurrent-switch_to_spare_wheel: validation suite correct 35->35, harmful 1->1, twin runs 124->108. R-sensor_disagree-mark_sensor_invalid: validation suite correct 35->35, harmful 1->1, twin runs 108->92. R-attitude_knowledge_lost-switch_attitude_sensor: validation suite correct 35->35, harmful 1->1, twin runs 92->80. R-solar_low-shed_loads: validation suite correct 35->35, harmful 1->1, twin runs 80->56. Rejected MUTANT-bias->heater: loses correct responses 35 -> 31; adds harmful responses 1 -> 5. Rejected MUTANT-wheel->safe: loses correct responses 35 -> 31. Rejected MUTANT-solar->safe: loses correct responses 35 -> 31.", | |
| "must": [ | |
| "admitted", | |
| "rule" | |
| ], | |
| "ok": true | |
| } | |
| ] | |
| }, | |
| "compute": { | |
| "mission_ticks": 320, | |
| "mission_seconds": 0.099, | |
| "ms_per_tick": 0.31, | |
| "max_decision_ms": 23.395, | |
| "twin_runs": 6, | |
| "peak_kib": 107 | |
| } | |
| } |