atravitz.bsky.social
@atravitz.bsky.social
Reposted
I suspect that fact that the vast majority of LLM users don’t seem to have received this (really very simple!) message is because the AI companies have a vested interest in us not understanding it. “ChatGPT is smart, it just makes mistakes sometimes” is much more marketable than the truth.
Chatbots — LLMs — do not know facts and are not designed to be able to accurately answer factual questions. They are designed to find and mimic patterns of words, probabilistically. When they’re “right” it’s because correct things are often written down, so those patterns are frequent. That’s all.
June 19, 2025 at 11:36 AM
I can personally recommend @ethanholz.com office hours💯
Hey computational scientists! I host monthly office hours and I am thinking of doing an intro to cloud compute for computational fields, is anyone interested? Thinking of going over some of the architecture considerations when building cloud pipelines in typical data flows in comp bio/comp chem.
June 17, 2025 at 5:05 PM
Reposted
on the flip side, now every time you see a straight person wearing jorts or beige you are legally obligated to tell them how much you like their corporate pride merch
May 30, 2025 at 10:09 PM
Reposted
If you weren't at @feworkshop.bsky.social in Boston last week, you missed Hannah Baumann giving the first sneak peek at results from our partner benchmarking study! Fortunately for you, Josh Horton wrote up a blog post about the early results: blog.omsf.io/the-free-ene...

#compChem
The Free Energy of Everything: Benchmarking OpenFE
Written by: Josh Horton, PhD Who is OpenFE and what do they do? Open Free Energy (OpenFE) sits at the nexus of academia and industry. We are developing an open-source software ecosystem for alchem...
blog.omsf.io
May 15, 2025 at 2:00 PM