Public KG record
Qwen3-8B
Recorded overview
Base model for performance evaluation.
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Benchmarks: in-distribution political refusal rate 84.75% with escalated-coverage
- Benchmarks: BANKING77：准确率0.747/0.803/0.807
- Benchmarks: BANKING77：ECE 0.240/0.184/0.095
- Benchmarks: 参与BANKING77基准测试
- Benchmarks: 原始读数选项反转率为0.230
- Benchmarks: 原始读数准确率为0.747
- Benchmarks: 原始读数ECE为0.240
- Benchmarks: 原始读数在5%错误率下可自动决策流量为7.7%
Evidence coverage
The KG currently records 13 known information dimensions and has 6 undocumented dimensions, linked to 17 public source articles. These counts describe record coverage, not model performance or verified fit.