csitfun/LogiEval

a benchmark suite for testing logical reasoning abilities of prompt-based models