Hi All!
In a technical report released on March 27, 2023, OpenAI provided a comprehensive brief on its most recent model, known as GPT-4. Included in this report were a set of exam results, which were visualized in the graphics attached. ---------
GREAT SCORES for GPT-4:
In the Law - Uniform Bar Exam category,
GPT-4 ranked percentile 90%
vs
GPT-3.5 ranked percentile: 10%
----------
VERY BAD SCORES::
BUT neither did well in the category: AP - English Language/Literature and Competitive Programming:
Advanced Placement (AP) - English Language.
GPT-4: 14% || GPT-3.5: 14%
Advanced Placement (AP) - English Literature
GPT4: 8% || GPT 3.5: 8%
Competitive Programming - Codeforces Rating
GPT-4: <5% || GPT-3.5: <5%
I thought I'd share. 😎
Have a great day!