Parcus Experiments in analyzing the correlation and effects of token budgets on a language model's ability to reason and generate accurate responses.
