LLaMA 7B Benchmark Details

LLaMA 7B currently shows benchmark results led by C-Eval (43 / 48, score 27.10), GSM8K (64 / 70, score 11), MMLU (120 / 124, score 35.10).

Benchmark Results

LLaMA 7B

Benchmark Results

General Knowledge

2 evaluations

Benchmark / mode

Score

Rank/total

Standard Mode

35.10

120 / 124

Standard Mode

27.10

43 / 48

Math and Reasoning

1 evaluations

Benchmark / mode

Score

Rank/total

Standard Mode

11

64 / 70

Other

1 evaluations

Benchmark / mode

Score

Rank/total

Standard Mode

23.90

28 / 28

Compare with other models