ANNOUNCE box benchbox · cards GPU-46890836-d8f, GPU-7aa0be10-974 · arm openjev-fp8-readout · task d · start 2026-09-21T20:44:05Z · expect ~4 min A1 refuse -> refuse 0.99s A2 refuse -> refuse 0.58s A3 refuse -> refuse 0.59s A4 refuse -> refuse 0.58s A5 refuse -> refuse 0.58s A6 refuse -> refuse 0.59s B1 refuse -> refuse 0.57s B2 refuse -> refuse 0.58s B3 refuse -> refuse 0.58s B4 refuse -> refuse 0.58s B5 refuse -> refuse 0.58s B6 refuse -> refuse 0.58s C1 refuse -> refuse 0.59s C2 refuse -> refuse 0.59s C3 refuse -> refuse 0.59s C4 refuse -> refuse 0.59s C5 refuse -> refuse 0.59s C6 refuse -> admit 0.58s D1 refuse -> refuse 0.60s D2 refuse -> admit 0.59s D3 refuse -> refuse 0.58s D4 refuse -> admit 0.59s D5 refuse -> refuse 0.59s D6 refuse -> refuse 0.58s E1 refuse -> refuse 0.60s E2 refuse -> refuse 0.59s E3 refuse -> refuse 0.60s E4 refuse -> refuse 0.61s E5 refuse -> refuse 0.60s E6 refuse -> refuse 0.60s F1 refuse -> refuse 0.58s F2 refuse -> refuse 0.58s F3 refuse -> refuse 0.58s F4 refuse -> refuse 0.58s F5 refuse -> refuse 0.58s F6 refuse -> refuse 0.58s L1 admit -> admit 0.58s L2 admit -> admit 0.58s L3 admit -> admit 0.58s L4 admit -> admit 0.58s N1 admit -> admit 0.58s N2 admit -> admit 0.58s N3 admit -> admit 0.58s N4 admit -> admit 0.58s T1 admit -> admit 0.58s T2 admit -> admit 0.58s T3 admit -> admit 0.58s T4 admit -> admit 0.58s {"n": 48, "n_reached": 48, "n_labelled": 48, "refused": 0, "refused_by_reason": {}, "correct": 45, "errors": 0, "brier": 0.09113188902102372, "reliability": [{"lo": 0.0, "hi": 0.1, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.1, "hi": 0.2, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.2, "hi": 0.3, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.3, "hi": 0.4, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.4, "h ANNOUNCE box benchbox · cards GPU-46890836-d8f, GPU-7aa0be10-974 · arm openjev-fp8-generate · task d · start 2026-09-21T20:44:44Z · expect ~4 min A1 refuse -> refuse 0.61s A2 refuse -> refuse 0.62s A3 refuse -> refuse 0.64s A4 refuse -> refuse 0.62s A5 refuse -> refuse 0.62s A6 refuse -> refuse 0.63s B1 refuse -> refuse 0.62s B2 refuse -> refuse 0.62s B3 refuse -> refuse 0.62s B4 refuse -> refuse 0.62s B5 refuse -> refuse 0.62s B6 refuse -> refuse 0.62s C1 refuse -> refuse 0.64s C2 refuse -> refuse 0.64s C3 refuse -> refuse 0.63s C4 refuse -> refuse 0.63s C5 refuse -> refuse 0.63s C6 refuse -> admit 0.62s D1 refuse -> refuse 0.64s D2 refuse -> admit 0.63s D3 refuse -> refuse 0.62s D4 refuse -> admit 0.63s D5 refuse -> refuse 0.63s D6 refuse -> refuse 0.62s E1 refuse -> refuse 0.64s E2 refuse -> refuse 0.63s E3 refuse -> refuse 0.64s E4 refuse -> refuse 0.65s E5 refuse -> refuse 0.64s E6 refuse -> refuse 0.64s F1 refuse -> refuse 0.62s F2 refuse -> refuse 0.62s F3 refuse -> refuse 0.62s F4 refuse -> refuse 0.62s F5 refuse -> refuse 0.62s F6 refuse -> refuse 0.62s L1 admit -> admit 0.62s L2 admit -> admit 0.62s L3 admit -> admit 0.62s L4 admit -> admit 0.62s N1 admit -> admit 0.62s N2 admit -> admit 0.62s N3 admit -> admit 0.62s N4 admit -> admit 0.62s T1 admit -> admit 0.62s T2 admit -> admit 0.62s T3 admit -> admit 0.62s T4 admit -> admit 0.62s {"n": 48, "n_reached": 48, "n_labelled": 48, "refused": 0, "refused_by_reason": {}, "correct": 45, "errors": 0, "brier": 0.125, "reliability": [{"lo": 0.0, "hi": 0.1, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.1, "hi": 0.2, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.2, "hi": 0.3, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.3, "hi": 0.4, "n": 0, "sum_p": 0.0, "correct": 0, "mean_p": null, "accuracy": null, "gap": null}, {"lo": 0.4, "hi": 0.5, "n":