C-Eval ↗38.9%
ChatGLM-6B · C-Eval · Accuracy · Zero-shot ↗
1 more reported result
34.5% · ChatGLM-6B · C-Eval · Accuracy · Five-shot · C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models ↗
Z.ai / MODEL EVIDENCE
ChatGLM-6B
2 benchmarks · 3 observations · 2 sources
Scores at a glance
2 benchmarks · large dot: ChatGLM-6B; small dots: closest models where reported
CMMLU ↗37.48%
ChatGLM-6B · CMMLU · Five-shot subject-macro accuracy · original five-shot comparison ↗
Closest models by score
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
LLaMA-65B →Meta
2 of 2 shared benchmark comparisons
2 of 2 shared benchmark comparisons
GPT-4 →OpenAI
2 of 2 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.
- CMMLU: Measuring massive multitask language understanding in Chinese — published result table ↗
All 8 authors
Li, Haonan, Zhang, Yixuan, Koto, Fajri, Yang, Yifei, Zhao, Hai, Gong, Yeyun, Duan, Nan, Baldwin, Timothy
- C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models — published result table ↗
All 13 authors
Yuzhen Huang, Yuzhuo Bai, Zhihao Zhu, Junlei Zhang, Jinghan Zhang, Tangjun Su, Junteng Liu, Chuancheng Lv, Yikai Zhang, Jiayi Lei, Yao Fu, Maosong Sun, Junxian He
Names used in source tables
ChatGLM-6B