AGIEval ↗29.8%
Phi-2 2.7b* · AGIEval · Accuracy · All · zero-shot ↗
Microsoft / MODEL EVIDENCE
Phi-2 2.7b*
4 benchmarks · 4 observations · 2 sources
Scores at a glance
4 benchmarks · large dot: Phi-2 2.7b*; small dots: closest models where reported
HumanEval+ ↗45.1%
phi-2-2.7B · HumanEval+ · Pass@1 · EvalPlus · official result data ↗
MBPP+ ↗54.2%
phi-2-2.7B · MBPP+ · Pass@1 · EvalPlus · official result data ↗
HumanEval ↗49.4%
phi-2-2.7B · HumanEval · Pass@1 · EvalPlus · official result data ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
3 of 4 shared benchmark comparisons
CodeLlama-34B →Meta
3 of 4 shared benchmark comparisons
3 of 4 shared benchmark comparisons
How similarity is measured
Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.
Sources
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
- Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation — published result table ↗
- AGIEval: A Human-Centric Benchmark for Evaluating Foundation Models — published result table ↗
All 9 authors
Wanjun Zhong, Ruixiang Cui, Yiduo Guo, Yaobo Liang, Shuai Lu, Yanlin Wang, Amin Saied, Weizhu Chen, Nan Duan
Names used in source tables
Phi-2 2.7b* · phi-2-2.7B