in the loop

Gemma 7B
#225 of 274 scored.

GoogleFeb 21, 2024Open weights

Capabilities Index

Range 102.4–116.5

112.0

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

No independent scores yet.

Epoch AI hasn't run Gemma 7B yet. Its scores show up here the morning after they do.

Everything else Epoch has for it

  • HellaSwag76.3%
  • TriviaQA72.3%
  • OpenBookQA71.5%
  • ARC AI271.1%
  • PIQA62.4%
  • Winogrande58.0%
  • MMLU54.8%
  • GSM8K46.4%
  • BBH40.1%
  • ANLI23.1%

On the index

Around it.

  1. #222Mistral 7B v0.1Mistral112.2
  2. #223Falcon-180BTII112.1
  3. #224internlm-20bShanghai AI Lab112.1
  4. #225Gemma 7BGoogle112.0
  5. #226DeepSeek LLM 67BDeepSeek110.5
  6. #227LLaMA-65BMeta110.2
  7. #228Falcon 2 11BTII109.6

Sources