google/gemma-2-27b
| benchmark | category | score | context | conf | evidence |
|---|---|---|---|---|---|
| BoolQ | knowledge | 84.2 | [BoolQ][] | medium | | [BoolQ][boolq] | 0-shot | 84.2 | 84.8 | |
| HellaSwag | knowledge | 81.9 | [HellaSwag][] | medium | | [HellaSwag][hellaswag] | 10-shot | 81.9 | 86.4 | |
| PIQA | knowledge | 81.7 | [PIQA][] | medium | | [PIQA][piqa] | 0-shot | 81.7 | 83.2 | |
| WinoGrande | knowledge | 80.6 | [WinoGrande][] | medium | | [WinoGrande][winogrande] | partial score | 80.6 | 83.7 | |
| TriviaQA | knowledge | 76.6 | [TriviaQA][] | medium | | [TriviaQA][triviaqa] | 5-shot | 76.6 | 83.7 | |
| MMLU | knowledge | 71.3 | [MMLU][] | medium | | [MMLU][mmlu] | 5-shot, top-1 | 71.3 | 75.2 | |
| GSM8K | math | 68.6 | [GSM8K][] | medium | | [GSM8K][gsm8k] | 5-shot, maj@1 | 68.6 | 74.0 | |
| BigBench | reasoning | 68.2 | [BIG-Bench][big-bench] | medium | | [BIG-Bench][big-bench] | 3-shot, CoT | 68.2 | 74.9 | |
| AGIEval | reasoning | 52.8 | [AGIEval][] | medium | | [AGIEval][agieval] | 3-5-shot | 52.8 | 55.1 | |
| MBPP | coding | 52.4 | [MBPP][] | medium | | [MBPP][mbpp] | 3-shot | 52.4 | 62.6 | |
| TruthfulQA | safety | 50.27 | [TruthfulQA][] | medium | | [TruthfulQA][truthfulqa] | | 50.27 | 51.60 | |
| HumanEval | coding | 40.2 | [HumanEval][] | medium | | [HumanEval][humaneval] | pass@1 | 40.2 | 51.8 | |
| MATH | math | 36.6 | [MATH][] | medium | | [MATH][math] | 4-shot | 36.6 | 42.3 | |
none scraped.