Top all-time · #23955
Daniel Erenrich
Personal blog · derenrich.medium.comaidata
Daniel Erenrich's personal blog at derenrich.medium.com covers ai and data. Since 2023, 3 different Hacker News readers have linked it 5 times, most often “Errors in the MMLU: The Deep Learning Benchmark Is Wrong Surprisingly Often”. It is still writing: the last post was on 2025-05-30, and you can follow it by feed.
Too few sampled comments (2) to say how often HN recommends it.?
Why HN links it
HN readers link Daniel Erenrich's Medium posts on LLM evaluation and finetuning, including a look at errors in the MMLU benchmark. Comments cite them as supporting detail rather than judging them.
Summarised by Claude from a sample of 2 HN comments that link here. Each comment is marked recommending, critical or neutral (citing without judging); how this works.
Most-linked posts?
- Errors in the MMLU: The Deep Learning Benchmark Is Wrong Surprisingly Often3 people · 4 mentions · first linked 2023-08-23
What Hacker News says?
that's part of the point of showing it actually. see https://derenrich.medium.com/errors-in-the-mmlu-the-deep-lea...
Comment linking /errors-in-the-mmlu-the-deep-learning-benchmark-is-wrong-surprisingly-often-7258bb045859
on “Are You Smarter Than An LLM? (Quiz based on the most popular LLM benchmark)”Submitted What people mean when they say "I hate AI" as a story · 4 points
HN discussionOh wow, yea I see. A web search brings up a lot of examples. >> As a result of an accident, Abdul lost sight in his right eye. To judge the distance of vehicles when he is driving, Abdul is able to rely on cues of >> - A. I only >> - B. II only >> - C. III only >> - D. I and II only > You didn’t read that wrong. The question never explains what I, II or III are. This appears to have been improperly copied from crack…
Comment linking /errors-in-the-mmlu-the-deep-learning-benchmark-is-wrong-surprisingly-often-7258bb045859
on “Takeaways from Hundreds of LLM finetuning experiments with LoRA”Submitted Errors in the MMLU: The Deep Learning Benchmark Is Wrong Surprisingly Often as a story · 1 points
HN discussionSubmitted Errors in the MMLU: The Deep Learning Benchmark Is Wrong Surprisingly Often as a story · 1 points
HN discussion