Author of "How Benchmaxxed is gpt-oss-120b?" in cmart.blog
How media typically covers Evan Rosenbaum
Evan Rosenbaum as author
OpenAI's gpt-oss-120b model ranks 15 positions lower on contamination-resistant LiveBench than on public benchmarks, suggesting potential training data leakage or overfitting to public benchmark questions.
“Author of "How Benchmaxxed is gpt-oss-120b?" in cmart.blog”