Gemma 2B
Open Sourcevon Google DeepMind · Veroeffentlicht 2024-01-01
29.1
Durchschn. Score
N/A
Eingabepreis
N/A
Ausgabepreis
N/A
Kontextfenster
text
Typ
Tested on 16 benchmarks with 29.1% average. Top scores: OpenBookQA (71.5%), HellaSwag (61.9%), PIQA (54.6%).
Benchmark-Ergebnisse
| Benchmark | Kategorie | Score | Bar |
|---|---|---|---|
| OpenBookQA | knowledge | 71.5 | |
| HellaSwag | knowledge | 61.9 | |
| PIQA | knowledge | 54.6 | |
| TriviaQA | knowledge | 53.2 | |
| Winogrande | knowledge | 30.8 | |
| IFEval | language | 26.6 | |
| MMLU | knowledge | 23.1 | |
| ANLI | knowledge | 23.1 | |
| ARC AI2 | knowledge | 22.8 | |
| MMLU-PRO | knowledge | 21.6 | |
| BBH (HuggingFace) | general | 21.1 | |
| GSM8K | math | 17.7 | |
| BBH | reasoning | 13.6 | |
| MUSR | reasoning | 11.0 | |
| MATH Level 5 | math | 7.4 | |
| GPQA | knowledge | 4.9 |