forked from third-party-mirrors/ollama
* llm: limit generation to 10x context size to avoid run on generations * add comment * simplify condition statement
* llm: limit generation to 10x context size to avoid run on generations * add comment * simplify condition statement