LessWrong 2.0 Reader

View: New · Old · Top

Restrict date range: Today · This week · This month · Last three months · This year · All time

← previous page (newer posts) · next page (older posts) →

On Anthropic’s Sleeper Agents Paper
Zvi · 2024-01-17T16:10:05.145Z · comments (5)
[link] Unlocking Solutions—By Understanding Coordination Problems
James Stephen Brown (james-brown) · 2024-07-27T04:52:13.435Z · comments (4)
AI #44: Copyright Confrontation
Zvi · 2023-12-28T14:30:10.237Z · comments (13)
Monthly Roundup #17: April 2024
Zvi · 2024-04-15T12:10:03.126Z · comments (4)
[link] Gwern: Why So Few Matt Levines?
kave · 2024-10-29T01:07:27.564Z · comments (8)
Towards a formalization of the agent structure problem
Alex_Altair · 2024-04-29T20:28:15.190Z · comments (5)
Math-to-English Cheat Sheet
nahoj · 2024-04-08T09:19:40.814Z · comments (5)
[link] On the Role of Proto-Languages
adamShimi · 2024-09-22T16:50:34.720Z · comments (1)
[link] OpenAI releases GPT-4o, natively interfacing with text, voice and vision
Martín Soto (martinsq) · 2024-05-13T18:50:52.337Z · comments (23)
Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
leogao · 2023-12-16T05:39:10.558Z · comments (5)
[link] Come to Manifest 2024 (June 7-9 in Berkeley)
Saul Munn (saul-munn) · 2024-03-27T21:30:17.306Z · comments (2)
[link] Land Reclamation is in the 9th Circle of Stagnation Hell
Maxwell Tabarrok (maxwell-tabarrok) · 2024-01-12T13:36:27.159Z · comments (6)
Safe Stasis Fallacy
Davidmanheim · 2024-02-05T10:54:44.061Z · comments (2)
[link] Google Gemini Announced
Jacob G-W (g-w1) · 2023-12-06T16:14:07.192Z · comments (22)
[Closed] PIBBSS is hiring in a variety of roles (alignment research and incubation program)
Nora_Ammann · 2024-04-09T08:12:59.241Z · comments (0)
[link] [Closed] Agent Foundations track in MATS
Vanessa Kosoy (vanessa-kosoy) · 2023-10-31T08:12:50.482Z · comments (1)
Thiel on AI & Racing with China
Ben Pace (Benito) · 2024-08-20T03:19:18.966Z · comments (10)
[link] the micro-fulfillment cambrian explosion
bhauth · 2023-12-04T01:15:34.342Z · comments (5)
On “first critical tries” in AI alignment
Joe Carlsmith (joekc) · 2024-06-05T00:19:02.814Z · comments (8)
AI #76: Six Shorts Stories About OpenAI
Zvi · 2024-08-08T13:50:04.659Z · comments (10)
Fat Tails Discourage Compromise
niplav · 2024-06-17T09:39:16.489Z · comments (5)
AMA: Earning to Give
jefftk (jkaufman) · 2023-11-07T16:20:10.972Z · comments (8)
AI #40: A Vision from Vitalik
Zvi · 2023-11-30T17:30:08.350Z · comments (12)
[link] LLMs seem (relatively) safe
JustisMills · 2024-04-25T22:13:06.221Z · comments (24)
Trading off Lives
jefftk (jkaufman) · 2024-01-03T03:40:05.603Z · comments (12)
[link] S-Risks: Fates Worse Than Extinction
aggliu · 2024-05-04T15:30:36.666Z · comments (2)
[question] Can we get an AI to "do our alignment homework for us"?
Chris_Leong · 2024-02-26T07:56:22.320Z · answers+comments (33)
Acting Wholesomely
owencb · 2024-02-26T21:49:16.526Z · comments (64)
Zvi's Manifold Markets House Rules
Zvi · 2023-11-13T00:28:02.147Z · comments (6)
We are headed into an extreme compute overhang
devrandom · 2024-04-26T21:38:21.694Z · comments (33)
[Intuitive self-models] 5. Dissociative Identity (Multiple Personality) Disorder
Steven Byrnes (steve2152) · 2024-10-15T13:31:46.157Z · comments (6)
Causal Graphs of GPT-2-Small's Residual Stream
David Udell · 2024-07-09T22:06:55.775Z · comments (7)
Self-Blinded L-Theanine RCT
niplav · 2023-10-31T15:24:57.717Z · comments (12)
[link] Open Phil releases RFPs on LLM Benchmarks and Forecasting
LawrenceC (LawChan) · 2023-11-11T03:01:09.526Z · comments (0)
2022 (and All Time) Posts by Pingback Count
Raemon · 2023-12-16T21:17:00.572Z · comments (14)
Be More Katja
Nathan Young · 2024-03-11T21:12:14.249Z · comments (0)
AI #37: Moving Too Fast
Zvi · 2023-11-09T17:50:04.324Z · comments (5)
Calendar feature geometry in GPT-2 layer 8 residual stream SAEs
Patrick Leask (patrickleask) · 2024-08-17T01:16:53.764Z · comments (0)
AI #50: The Most Dangerous Thing
Zvi · 2024-02-08T14:30:13.168Z · comments (4)
Per protocol analysis as medical malpractice
braces · 2024-01-31T16:22:21.367Z · comments (8)
[link] Breaking Circuit Breakers
mikes · 2024-07-14T18:57:20.251Z · comments (13)
AI #71: Farewell to Chevron
Zvi · 2024-07-04T13:40:05.905Z · comments (9)
BatchTopK: A Simple Improvement for TopK-SAEs
Bart Bussmann (Stuckwork) · 2024-07-20T02:20:51.848Z · comments (0)
The Stochastic Parrot Hypothesis is debatable for the last generation of LLMs
Quentin FEUILLADE--MONTIXI (quentin-feuillade-montixi) · 2023-11-07T16:12:20.031Z · comments (20)
Reflections on my first year of AI safety research
Jay Bailey · 2024-01-08T07:49:08.147Z · comments (3)
AI #45: To Be Determined
Zvi · 2024-01-04T15:00:05.936Z · comments (4)
Parental Writing Selection Bias
jefftk (jkaufman) · 2024-10-13T14:00:03.225Z · comments (3)
Announcing the Double Crux Bot
sanyer (santeri-koivula) · 2024-01-09T18:54:15.361Z · comments (8)
A D&D.Sci Dodecalogue
abstractapplic · 2024-04-12T01:10:01.625Z · comments (0)
Can we build a better Public Doublecrux?
Raemon · 2024-05-11T19:21:53.326Z · comments (6)
← previous page (newer posts) · next page (older posts) →