LLMs can process massive contexts (1M+ tokens), reasoning like a human, but they can’t see individual letters due to tokenization.
I always try to give clear prompts with examples and step-by-step instructions. This helped me always getting the most out of it.
out it. instructions. (1M+ due massive This see helped try with the prompts and tokens), process can human, most LLMs getting but step-by-step
I contexts can’t always of me to letters individual reasoning always to tokenization. like they a examples give clear
