LLaMA2 13B Benchmark Details

LLaMA2 13B currently shows benchmark results led by AGIEval (17 / 28, score 39.10), MMLU (100 / 124, score 54.84), GSM8K (58 / 70, score 28.70).

Benchmark Results

LLaMA2 13B

Benchmark Results

General Knowledge

1 evaluations

Benchmark / mode

Score

Rank/total

MMLU

Standard Mode

54.84

100 / 124

Math and Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

GSM8K

Standard Mode

28.70

58 / 70

Coding and Software Engineer

2 evaluations

Benchmark / mode

Score

Rank/total

MBPP

Standard Mode

27.60

64 / 70

HumanEval

Standard Mode

20.10

93 / 101

Other

1 evaluations

Benchmark / mode

Score

Rank/total

AGIEval

Standard Mode

39.10

17 / 28

Other

4 evaluations

Benchmark / mode

Score

Rank/total

MBPP pass@100

Standard Mode

69.50

12 / 13

HumanEval pass@100

Standard Mode

61.20

19 / 21

MBPP pass@10

Standard Mode

48.10

12 / 13

HumanEval pass@10

Standard Mode

34.80

17 / 18

Compare with other models