Aggregated
Benchmarks
TaskDatasetSOTA resultTrendResults
Aggregated MMLU, GSM8K, HumanEval
88.33Average Accuracy
42
Aggregated MMLU, BoolQ, OpenBookQA, RTE
70.4Average Accuracy
42
Aggregated Average Score
62.7Average Score
36
Aggregated
71.44Accuracy
28
Aggregated AL, CH, CO, EY, GE, HE, HI, HO, JA, MI, OT, YE
2.17Rank
17
Aggregated All Benchmarks
40.3Average Score
12
Aggregated GSM8K, TruthfulQA, TriviaQA, CNN/DM, MMLU
48.6Average Score
9
Aggregated MMLU, HellaSwag, TruthfulQA, GSM8K, MATH, MBPP, HumanEval
40.35Average Score
9
Aggregated NQ, TQA, HQA, 2Wiki, Musique
34EM
8
Aggregated
0.76InfoM
8
Aggregated Univariate, 1D, and 2D regression tasks (test)
1.78SSE
5
Aggregated (Shapes3D, MPI3D, Falcor3D, Isaac3D)
0.65InfoM Score
5
Aggregated (Sudoku, Maze, ARC, DDE)
3.04Fp Score
4
Aggregated SST, Ev.Inf, AG, and M.RC
0.525Alpha Score
4
Aggregated Mean All & Mean R1M (test)
0.601Mean All
2