DataLearner logo

Phi-1.5 Benchmark Details

Phi-1.5 currently shows benchmark results led by HumanEval (74 / 101, score 34.10), GSM8K (51 / 68, score 40.20), MBPP (58 / 69, score 37.70).

Benchmark Results

Phi-1.5

Benchmark Results

Thinking

Knowledge Exams

1 evaluations
Benchmark / mode
Score
Rank/total
MMLU
Standard Mode
37.60
118 / 124

Mathematics

1 evaluations
Benchmark / mode
Score
Rank/total
GSM8K
Standard Mode
40.20
51 / 68

Code Generation & Editing

2 evaluations
Benchmark / mode
Score
Rank/total
MBPP
Standard Mode
37.70
58 / 69
HumanEval
Standard Mode
34.10
74 / 101