请问有什么好的办法解决token消耗量大的问题吗？

审计一个gitlab项目，使用了847k 的token，感觉使用成本太高了，请问有更好的解决办法吗？

还有，这里显示的125行代码是否有误，实际整个项目不止这么多行代码的

Start by reproducing the GitLab project audit described in the issue and compare the reported 847k-token usage with the displayed 125 lines of code. Determine whether the line count is incorrect and identify a concrete, verifiable way to reduce token consumption; the issue provides no file or test entry point.

Written by the indexing model from the issue text.