-
-
-
-
-
Simple program that takes in raw data and outputs data compatible with GPT-4 evaluation. Specifically, for evaluations testing mono-alphabetic substitution cipher.
Python UpdatedApr 6, 2023 -
evals Public
Forked from openai/evalsEvals is a framework for evaluating OpenAI models and an open-source registry of benchmarks.
Python MIT License UpdatedApr 6, 2023