This is not completely true, though. We expected GPT and LLMs in general to be bad in tasks that are not language-focused, exactly because they are large LANGUA

This is not completely true, though. We expected GPT and LLMs in general to be bad in tasks that are not language-focused, exactly because they are large LANGUAGE models. Mathematics is one of them. Though it is also true that one may not have expected that GPT would be so bad as it is in extrapolating, nobody should have expected it to any good in any involved mathematical argumentation.