clp-research / clembench

A Framework for the Systematic Evaluation of Chat-Optimized Language Models as Conversational Agents and an Extensible Benchmark
MIT License
19 stars 26 forks source link

[wordle] fix re-prompt prompts #98

Open davidschlangen opened 3 days ago

davidschlangen commented 3 days ago

Double check all prompts. There is at least one that needs to be fixed “the word should only contain alphabets”