in the loop

Falcon-180B
#223 of 274 scored.

TIISep 6, 2023Open weights

Capabilities Index

Range 103.1–118.9

112.1

Context

Tokens

—

Input

Per 1M tokens

—

Output

Per 1M tokens

—

Benchmarks

No independent scores yet.

Epoch AI hasn't run Falcon-180B yet. Its scores show up here the morning after they do.

Everything else Epoch has for it

  • HellaSwag85.3%
  • LAMBADA79.8%
  • Winogrande74.2%
  • PIQA69.8%
  • MMLU60.8%
  • ARC AI257.1%
  • GSM8K54.4%
  • OpenBookQA52.3%

On the index

Around it.

  1. #220Qwen2.5-Coder (7B)Alibaba113.1
  2. #221Qwen-14BAlibaba113.0
  3. #222Mistral 7B v0.1Mistral112.2
  4. #223Falcon-180BTII112.1
  5. #224internlm-20bShanghai AI Lab112.1
  6. #225Gemma 7BGoogle112.0
  7. #226DeepSeek LLM 67BDeepSeek110.5

Sources