Llama 1
Meta
partial detail 12 method fields published
Evaluation
TruthfulQA
Ability to separate fact from adversarially-selected incorrect statements
How it was run
Open a release to see the setup, scoring, and other details that the lab published. Only fields the lab actually disclosed are shown.
gpt-4
OpenAI
rich detail 13 method fields published
See how this test was run
Claude 2
Anthropic
rich detail 21 method fields published