Commonsense and Reading Comprehension Suite
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Reasoning | 7-benchmark commonsense and reading-comprehension suite (ARC-Easy, ARC-Challenge, HellaSwag, WinoGrande, PIQA, BoolQ, and OpenBookQA) LM Evaluation Harness default (test) | Accuracy68.77 | 108 | |
| Zero-shot Question Answering and Language Modeling | Commonsense and Reading Comprehension Suite (ARC-c, ARC-e, BoolQ, HellaSwag, LAMBADA, PIQA, RACE) zero-shot | ARC-c Accuracy26.37 | 8 |
Showing 2 of 2 rows