{ "arc_challenge": { "n": 1172, "accuracy": 0.878839590443686, "nll": 0.3434354264088279, "brier": 0.18004018282065612, "ece": 0.018640702693919233, "latency_p50_ms": 173.45179200128769, "latency_p95_ms": 241.31601209974178, "mean_confidence": 0.8609673274686053 }, "winogrande": { "n": 1267, "accuracy": 0.664561957379637, "nll": 0.8732282598352692, "brier": 0.5079363220206212, "ece": 0.1986676781891346, "latency_p50_ms": 168.07529400102794, "latency_p95_ms": 241.4132092009823, "mean_confidence": 0.8493462892926714 }, "arc_easy": { "n": 2376, "accuracy": 0.9562289562289562, "nll": 0.15331558248276225, "brier": 0.07269179989034444, "ece": 0.029571857402955024, "latency_p50_ms": 172.40582649901626, "latency_p95_ms": 231.58336125015921, "mean_confidence": 0.9269831239375699 }, "hellaswag": { "n": 10042, "accuracy": 0.8977295359490142, "nll": 0.2736215284007136, "brier": 0.14571887516017584, "ece": 0.012820075397336044, "latency_p50_ms": 230.12878749796073, "latency_p95_ms": 276.0439652007335, "mean_confidence": 0.8883940220236488 }, "mmlu": { "n": 14042, "accuracy": 0.6441390115368181, "nll": 0.8705675106095903, "brier": 0.45396635565340543, "ece": 0.03314363695835977, "latency_p50_ms": 196.04511249963252, "latency_p95_ms": 303.28189720094076, "mean_confidence": 0.6772229938346321 } }