3 ms·
You could run it through thrice with a different prompt/temperature/model and pick the majority result (or exit with success on the first two passing runs).
by ukuina 3y ago
You could run it through thrice with a different prompt/temperature/model and pick the majority result (or exit with success on the first two passing runs).
- akoboldfrying 3y agoGood idea. If the data is a list of records where the order isn't important, randomly permuting them (ETA: then sorting the final outputs) would be another option. ETA2: Would the downvoter care to explain why? Genuinely puzzled.