Nina Panickssery

Karma: 2,486

https://ninapanickssery.com/

Views purely my own unless clearly stated otherwise

Breastfeeding and IQ: Effects shrink as you control for more confounders

Nina Panickssery25 Aug 2025 18:43 UTC

39 points

2 comments1 min readLW link

(blog.ninapanickssery.com)

Interiors can be more fun

Nina Panickssery13 Aug 2025 22:42 UTC

34 points

6 comments4 min readLW link

(blog.ninapanickssery.com)

Negative utilitarianism is more intuitive than you think

Nina Panickssery11 Aug 2025 16:13 UTC

13 points

25 comments3 min readLW link

(blog.ninapanickssery.com)

[Fiction] Our Trial

Nina Panickssery21 Jul 2025 3:56 UTC

68 points

1 comment3 min readLW link

(ninapanickssery.substack.com)

Why do LLMs hallucinate?

Nina Panickssery13 Jul 2025 0:09 UTC

24 points

1 comment5 min readLW link

(ninapanickssery.substack.com)

Surprises and learnings from almost two months of Leo Panickssery

Nina Panickssery12 Jul 2025 23:33 UTC

208 points

12 comments6 min readLW link

(ninapanickssery.substack.com)

[Question] How does the LessWrong team generate the website illustrations?

Nina Panickssery23 Jun 2025 0:05 UTC

13 points

1 comment1 min readLW link

My favorite Soviet songs

Nina Panickssery15 Jun 2025 2:48 UTC

22 points

1 comment5 min readLW link

(ninapanickssery.substack.com)

Why I am not a successionist

Nina Panickssery4 May 2025 19:08 UTC

67 points

54 comments2 min readLW link

(ninapanickssery.substack.com)

Role embeddings: making authorship more salient to LLMs

Nina Panickssery and Christopher Ackerman

7 Jan 2025 20:13 UTC

50 points

0 comments8 min readLW link

Nina Panickssery’s Shortform

Nina Panickssery7 Jan 2025 2:06 UTC

7 points

51 comments LW link

Investigating the Ability of LLMs to Recognize Their Own Writing

Christopher Ackerman and Nina Panickssery

30 Jul 2024 15:41 UTC

32 points

0 comments15 min readLW link

Jailbreak steering generalization

Sarah Ball and Nina Panickssery

20 Jun 2024 17:25 UTC

41 points

4 comments2 min readLW link

(arxiv.org)

Soviet comedy film recommendations

Nina Panickssery9 Jun 2024 23:40 UTC

42 points

11 comments2 min readLW link

(open.substack.com)

Steering Llama-2 with contrastive activation additions

Nina Panickssery, Wuschel Schulz, NickGabs, Meg, evhub and TurnTrout

2 Jan 2024 0:47 UTC

125 points

29 comments8 min readLW link

(arxiv.org)

Comparing representation vectors between llama 2 base and chat

Nina Panickssery28 Oct 2023 22:54 UTC

36 points

5 comments2 min readLW link

Investigating the learning coefficient of modular addition: hackathon project

Nina Panickssery and Dmitry Vaintrob

17 Oct 2023 19:51 UTC

94 points

5 comments12 min readLW link

Influence functions—why, what and how

Nina Panickssery15 Sep 2023 20:42 UTC

75 points

6 comments8 min readLW link

Red-teaming language models via activation engineering

Nina Panickssery26 Aug 2023 5:52 UTC

69 points

6 comments9 min readLW link

The Low-Hanging Fruit Prior and sloped valleys in the loss landscape

Dmitry Vaintrob and Nina Panickssery

23 Aug 2023 21:12 UTC

82 points

1 comment13 min readLW link