{ "tasks": { "hellaswag": { "n": 10042, "seconds": 146.01327709999168, "routes": { "cached/language+evidence": 2973, "cached/language": 7068, "cached/exact-arithmetic": 1 }, "changed_system_predictions": 0, "fresh_sql_parity_checks": 3, "maximum_likelihood_error": 0.0, "prediction_sha256": "ae5e8d2fb43e21a1810c4e8518e5a54256cf7ea66b77a273f1229b753360c7c9", "primary": "char", "native_correct": 2809, "native_percent": 27.972515435172276, "previous_native_correct": 2809, "previous_native_percent": 27.972515435172276, "native_change_points": 0.0, "assisted_correct": 2805, "assisted_percent": 27.9326827325234, "previous_assisted_percent": 27.93, "assisted_change_points": 0.0 }, "arc_easy": { "n": 2376, "seconds": 22.353922400041483, "routes": { "cached/language+evidence": 2374, "cached/exact-arithmetic": 1, "cached/language": 1 }, "changed_system_predictions": 0, "fresh_sql_parity_checks": 3, "maximum_likelihood_error": 0.0, "prediction_sha256": "f9b4a6ca075ad0110315bb1dd19e8eac4b6df47252730596b25c6933dae521d9", "primary": "char", "native_correct": 632, "native_percent": 26.5993265993266, "previous_native_correct": 632, "previous_native_percent": 26.5993265993266, "native_change_points": 0.0, "assisted_correct": 1476, "assisted_percent": 62.121212121212125, "previous_assisted_percent": 62.12, "assisted_change_points": 0.0 }, "arc_challenge": { "n": 1172, "seconds": 11.865461300010793, "routes": { "cached/language+evidence": 1169, "cached/exact-arithmetic": 2, "cached/language": 1 }, "changed_system_predictions": 0, "fresh_sql_parity_checks": 3, "maximum_likelihood_error": 0.0, "prediction_sha256": "301b66946ce1cda420019aca3e1240bc9c262f2cb4f73e909b3ad19e61f76675", "primary": "char", "native_correct": 260, "native_percent": 22.18430034129693, "previous_native_correct": 260, "previous_native_percent": 22.18430034129693, "native_change_points": 0.0, "assisted_correct": 371, "assisted_percent": 31.655290102389078, "previous_assisted_percent": 31.66, "assisted_change_points": -0.0 }, "piqa": { "n": 1838, "seconds": 10.459679999970831, "routes": { "cached/language": 433, "cached/language+evidence": 1405 }, "changed_system_predictions": 0, "fresh_sql_parity_checks": 3, "maximum_likelihood_error": 0.0, "prediction_sha256": "cf2a6f1adc18959b67dce5af1fa668e7f4301d7eb9b25d3918fdc221c8369597", "primary": "char", "native_correct": 899, "native_percent": 48.91186071817192, "previous_native_correct": 899, "previous_native_percent": 48.91186071817192, "native_change_points": 0.0, "assisted_correct": 1078, "assisted_percent": 58.65070729053319, "previous_assisted_percent": 58.65, "assisted_change_points": 0.0 }, "arithmark3": { "n": 1000, "seconds": 9.040633900032844, "routes": { "cached/exact-arithmetic": 680, "cached/language+evidence": 320 }, "changed_system_predictions": 0, "fresh_sql_parity_checks": 3, "maximum_likelihood_error": 0.0, "prediction_sha256": "9cb2c70adfac81b8fde8c1fa2079553530d667b15d3190db54a7b76dce26f83d", "primary": "token", "native_correct": 378, "native_percent": 37.8, "previous_native_correct": 378, "previous_native_percent": 37.8, "native_change_points": 0.0, "assisted_correct": 695, "assisted_percent": 69.5, "previous_assisted_percent": 69.5, "assisted_change_points": 0.0 } }, "total_questions": 16428, "seconds": 200.08493640000233, "native_mean_percent": 32.693600618793546, "assisted_mean_percent": 49.97197844933156, "current_index_formula": "Chance-normalized HellaSwag + combined ARC + PIQA + 0.65*ArithMark, divided by 3.65", "local_index_native": 3.3067104959787845, "local_index_assisted": 24.373331426529237, "whole_system_evaluation_cached": true, "fresh_likelihoods": true, "training": false, "official_submission": false, "freeze_sha256": "d1f3b91f49fe8036b10a814daca5eb20e9e7c64f7551e86cc475a8c14373b1e6", "sql_parity_sha256": "27cc34d0f40d3f369c42d6f17a9ef3b42bb94c0f514f21037af7be08235128c1" }