It's the whole answer being correct that's the important thing, and if you compare GPT 3 vs where we are today only 5 years later the progress in accuracy, knowledge and intelligence is jaw dropping.

▲

th0ma5 8 months ago | parent [-]

I have no idea what you're talking about because they still screw up in the exact same way as gpt3.

▲

Closi 8 months ago | parent [-]

The hallucination quantity and severity is way less in new frontier models.

	▲	th0ma5 8 months ago \| parent [-]
		But not more predictable or regular.