Adam Scherlis

Karma: 988

Estimating the Probability of Sampling a Trained Neural Network at Random

Adam Scherlis and Nora Belrose

Mar 1, 2025, 2:11 AM

32 points

10 comments1 min readLW link

(arxiv.org)

What can we learn about childrearing from J. S. Mill?

Adam ScherlisApr 2, 2024, 6:06 AM

10 points

1 comment1 min readLW link

Two Percolation Puzzles

Adam ScherlisJul 4, 2023, 5:34 AM

43 points

14 comments1 min readLW link

(adam.scherlis.com)

GPT-175bee

Adam Scherlis and LawrenceC

Feb 8, 2023, 6:58 PM

122 points

14 comments1 min readLW link

How to export Android Chrome tabs to an HTML file in Linux (as of February 2023)

Adam ScherlisFeb 2, 2023, 2:03 AM

7 points

3 comments2 min readLW link

(adam.scherlis.com)

Inner Misalignment in “Simulator” LLMs

Adam ScherlisJan 31, 2023, 8:33 AM

84 points

12 comments4 min readLW link

Fun math facts about 2023

Adam ScherlisJan 1, 2023, 11:38 PM

9 points

6 comments1 min readLW link

A hundredth of a bit of extra entropy

Adam ScherlisDec 24, 2022, 9:12 PM

84 points

4 comments3 min readLW link

An exploration of GPT-2′s embedding weights

Adam ScherlisDec 13, 2022, 12:46 AM

44 points

4 comments10 min readLW link

A brainteaser for language models

Adam ScherlisDec 12, 2022, 2:43 AM

47 points

3 comments2 min readLW link

Adam Scherlis’s Shortform

Adam ScherlisDec 9, 2022, 12:11 AM

4 points

16 comments LW link

New Frontiers in Mojibake

Adam ScherlisNov 26, 2022, 2:37 AM

60 points

7 comments6 min readLW link 1 review

(adam.scherlis.com)

Cryptic symbols

Adam ScherlisOct 28, 2022, 6:44 AM

6 points

17 comments1 min readLW link

(adam.scherlis.com)

Five views of Bayes’ Theorem

Adam ScherlisJul 2, 2022, 2:25 AM

38 points

4 comments1 min readLW link

One is (almost) normal in base π

Adam ScherlisJul 1, 2022, 4:05 AM

14 points

0 comments1 min readLW link

(adam.scherlis.com)

Understanding the two-head strategy for teaching ML to answer questions honestly

Adam ScherlisJan 11, 2022, 11:24 PM

29 points

1 comment10 min readLW link

Fun with bitstrings and bijections

Adam ScherlisJan 2, 2022, 5:07 AM

6 points

0 comments1 min readLW link

(adam.scherlis.com)

Gradient Hacking via Schelling Goals

Adam ScherlisDec 28, 2021, 8:38 PM

33 points

4 comments4 min readLW link

A Generalization of ROC AUC for Binary Classifiers

Adam ScherlisDec 4, 2021, 9:47 PM

10 points

0 comments2 min readLW link

(adam.scherlis.com)

What’s the weirdest way to win this game?

Adam ScherlisNov 21, 2021, 5:18 AM

9 points

5 comments1 min readLW link

(adam.scherlis.com)