{ "arc_easy": { "n": 2376, "accuracy": 0.9558080808080808, "nll": 0.22179265867232187, "brier": 0.07866352502747752, "ece": 0.03341212654329345, "latency_p50_ms": 153.7982515001204, "latency_p95_ms": 216.1515892476018, "mean_confidence": 0.9868527784304538 }, "arc_challenge": { "n": 1172, "accuracy": 0.8737201365187713, "nll": 0.6831067400476731, "brier": 0.21717192716282274, "ece": 0.0939787680494793, "latency_p50_ms": 158.1707515015296, "latency_p95_ms": 239.32236594882852, "mean_confidence": 0.9676989045682511 }, "winogrande": { "n": 1267, "accuracy": 0.601420678768745, "nll": 1.2788768116978686, "brier": 0.6614413444986983, "ece": 0.3052572014015491, "latency_p50_ms": 154.72152499933145, "latency_p95_ms": 223.75241669906248, "mean_confidence": 0.9024290672675535 }, "hellaswag": { "n": 10042, "accuracy": 0.7650866361282613, "nll": 1.1006902005839516, "brier": 0.3895498661609883, "ece": 0.16545367857507678, "latency_p50_ms": 259.9183570000605, "latency_p95_ms": 484.0487622521322, "mean_confidence": 0.9305298846809178 }, "mmlu": { "n": 14042, "accuracy": 0.6618715282723259, "nll": 1.6791403682676391, "brier": 0.5577819683925032, "ece": 0.24181974065906128, "latency_p50_ms": 176.33997150187497, "latency_p95_ms": 324.31165310117643, "mean_confidence": 0.9036912689313833 } }