RSS

Subhash Kantamneni

Karma: 503

Nat­u­ral Lan­guage Au­toen­coders Pro­duce Un­su­per­vised Ex­pla­na­tions of LLM Activations

Subhash Kantamneni, kitft, Euan Ong and Sam Marks
7 May 2026 20:21 UTC
213 points
35 comments8 min readLW link

Ac­ti­va­tion Or­a­cles: Train­ing and Eval­u­at­ing LLMs as Gen­eral-Pur­pose Ac­ti­va­tion Explainers

Sam Marks, Adam Karvonen, James Chua, Subhash Kantamneni, Euan Ong, Julian Minder, Clément Dumas and Owain_Evans
18 Dec 2025 20:21 UTC
154 points
11 comments8 min readLW link
(arxiv.org)

Scal­ing Laws for Scal­able Oversight

Subhash Kantamneni, Josh Engels, David Baek and Max Tegmark
30 Apr 2025 12:13 UTC
38 points
1 comment9 min readLW link

Take­aways From Our Re­cent Work on SAE Probing

Josh Engels, Subhash Kantamneni, Senthooran Rajamanoharan and Neel Nanda
3 Mar 2025 19:50 UTC
30 points
4 comments5 min readLW link

Lan­guage Models Use Tri­gonom­e­try to Do Addition

Subhash Kantamneni5 Feb 2025 13:50 UTC
80 points
1 comment10 min readLW link

SAE Prob­ing: What is it good for?

Subhash Kantamneni, Josh Engels, Senthooran Rajamanoharan and Neel Nanda
1 Nov 2024 19:23 UTC
34 points
0 comments11 min readLW link