Joseph Miller

Karma: 2,481

60 U.K. Lawmakers Accuse Google of Breaking AI Safety Pledge

Joseph Miller29 Aug 2025 16:09 UTC

50 points

1 comment1 min readLW link

(time.com)

Dreams of Ideas

Joseph Miller19 May 2025 2:15 UTC

90 points

3 comments4 min readLW link

Gradient Routing: Masking Gradients to Localize Computation in Neural Networks

cloud, Jacob G-W, Evzen, Joseph Miller and TurnTrout

6 Dec 2024 22:19 UTC

169 points

12 comments11 min readLW link

(arxiv.org)

Transformer Circuit Faithfulness Metrics Are Not Robust

Joseph Miller, bilalchughtai and William_S

12 Jul 2024 3:47 UTC

104 points

5 comments7 min readLW link

(arxiv.org)

Joseph Miller’s Shortform

Joseph Miller21 May 2024 20:50 UTC

5 points

59 comments LW link

How To Do Patching Fast

Joseph Miller11 May 2024 20:13 UTC

44 points

8 comments4 min readLW link

Why I’m doing PauseAI

Joseph Miller30 Apr 2024 16:21 UTC

104 points

16 comments4 min readLW link

Global Pause AI Protest 10/21

Holly_Elmore, Joseph Miller and joepio

14 Oct 2023 3:20 UTC

5 points

0 comments1 min readLW link

The International PauseAI Protest: Activism under uncertainty

Joseph Miller12 Oct 2023 17:36 UTC

32 points

1 comment4 min readLW link

Even Superhuman Go AIs Have Surprising Failure Modes

AdamGleave, EuanMcLean, Tony Wang, Kellin Pelrine, Tom Tseng, Yawen Duan, Joseph Miller and MichaelDennis

20 Jul 2023 17:31 UTC

130 points

22 comments10 min readLW link

(far.ai)

We Found An Neuron in GPT-2

Joseph Miller and Clement Neo

11 Feb 2023 18:27 UTC

143 points

23 comments7 min readLW link

(clementneo.com)