4 ms·
It's just a guess, based on the LLM never getting writers block and being able to hold a good amount of context in mind for every output token.
by sp332 3y ago
It's just a guess, based on the LLM never getting writers block and being able to hold a good amount of context in mind for every output token.