4 ms·
The LLM probably doesn't have enough sample texts where the response is requested to be character limited to have generalised that as a constraint.
by simonh 3y ago
The LLM probably doesn't have enough sample texts where the response is requested to be character limited to have generalised that as a constraint.
- pixl97 3y agoYa, you'd have to adjust your initial prompt to GPT to remove as much prose and limit responses to be as short and terse as possible. Add to that GPT doesn't count characters, but instead tokens and it's difficult.