in the loop

Cerebras-GPT-13B
#272 of 274 scored.

CerebrasMar 20, 2023Open weights

Capabilities Index

Range 66.0–92.0

83.2

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

No independent scores yet.

Epoch AI hasn't run Cerebras-GPT-13B yet. Its scores show up here the morning after they do.

Everything else Epoch has for it

  • PIQA47.0%
  • HellaSwag45.9%
  • Winogrande21.6%
  • OpenBookQA14.4%
  • ARC AI29.9%
  • MMLU1.6%

On the index

Around it.

  1. #269DeepSeek Coder 6.7BDeepSeek89.6
  2. #270StarCoder 2 3BHugging Face88.8
  3. #271Qwen2.5-Coder-0.5BAlibaba88.2
  4. #272Cerebras-GPT-13BCerebras83.2
  5. #273DeepSeek Coder 1.3BDeepSeek63.6
  6. #274stablelm-tuned-alpha-7bStability AI55.8

Sources