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) →

flowing like water; hard like stone
lsusr · 2024-02-20T03:20:46.531Z · comments (4)
Deceptive agents can collude to hide dangerous features in SAEs
Simon Lermen (dalasnoin) · 2024-07-15T17:07:33.283Z · comments (0)
[link] Evaluating Synthetic Activations composed of SAE Latents in GPT-2
Giorgi Giglemiani (Rakh) · 2024-09-25T20:37:48.227Z · comments (0)
Incentive Learning vs Dead Sea Salt Experiment
Steven Byrnes (steve2152) · 2024-06-25T17:49:01.488Z · comments (1)
Cheap Whiteboards!
Johannes C. Mayer (johannes-c-mayer) · 2024-08-08T13:52:59.627Z · comments (2)
Interpretability of SAE Features Representing Check in ChessGPT
Jonathan Kutasov (jonathan-kutasov) · 2024-10-05T20:43:36.679Z · comments (2)
[question] Supposing the 1bit LLM paper pans out
O O (o-o) · 2024-02-29T05:31:24.158Z · answers+comments (11)
European Progress Conference
Martin Sustrik (sustrik) · 2024-10-06T11:10:03.819Z · comments (11)
On the 2nd CWT with Jonathan Haidt
Zvi · 2024-04-05T17:30:05.223Z · comments (3)
NYU Code Debates Update/Postmortem
David Rein (david-rein) · 2024-05-24T16:08:06.151Z · comments (4)
There aren't enough smart people in biology doing something boring
Abhishaike Mahajan (abhishaike-mahajan) · 2024-10-21T15:52:04.482Z · comments (13)
The Foraging (Ex-)Bandit [Ruleset & Reflections]
abstractapplic · 2024-11-14T20:16:21.535Z · comments (3)
Probably Not a Ghost Story
George Ingebretsen (george-ingebretsen) · 2024-06-12T22:55:26.264Z · comments (4)
[question] Why do Minimal Bayes Nets often correspond to Causal Models of Reality?
Dalcy (Darcy) · 2024-08-03T12:39:44.085Z · answers+comments (1)
Ackshually, many worlds is wrong
tailcalled · 2024-04-11T20:23:59.416Z · comments (42)
[link] A brief history of the automated corporation
owencb · 2024-11-04T14:35:04.906Z · comments (1)
[link] Secret US natsec project with intel revealed
Nathan Helm-Burger (nathan-helm-burger) · 2024-05-25T04:22:11.624Z · comments (0)
Option control
Joe Carlsmith (joekc) · 2024-11-04T17:54:03.073Z · comments (0)
Meetup In a Box: Year In Review
Czynski (JacobKopczynski) · 2024-02-14T01:18:28.259Z · comments (0)
Just because an LLM said it doesn't mean it's true: an illustrative example
dirk (abandon) · 2024-08-21T21:05:59.691Z · comments (12)
[link] Let's Design A School, Part 2.1 School as Education - Structure
Sable · 2024-05-02T22:04:30.435Z · comments (2)
[link] Death notes - 7 thoughts on death
Nathan Young · 2024-10-28T15:01:13.532Z · comments (1)
Vote in the LessWrong review! (LW 2022 Review voting phase)
habryka (habryka4) · 2024-01-17T07:22:17.921Z · comments (9)
Chat Bankman-Fried: an Exploration of LLM Alignment in Finance
claudia.biancotti · 2024-11-18T09:38:35.723Z · comments (4)
Smartphone Etiquette: Suggestions for Social Interactions
Declan Molony (declan-molony) · 2024-06-04T06:01:03.336Z · comments (4)
Essaying Other Plans
Screwtape · 2024-03-06T22:59:06.240Z · comments (4)
[link] Care Doesn't Scale
stavros · 2024-10-28T11:57:38.742Z · comments (1)
$250K in Prizes: SafeBench Competition Announcement
ozhang (oliver-zhang) · 2024-04-03T22:07:41.171Z · comments (0)
[question] Seeking AI Alignment Tutor/Advisor: $100–150/hr
MrThink (ViktorThink) · 2024-10-05T21:28:16.491Z · answers+comments (3)
[link] Can a Bayesian Oracle Prevent Harm from an Agent? (Bengio et al. 2024)
mattmacdermott · 2024-09-01T07:46:26.647Z · comments (0)
SAEs you can See: Applying Sparse Autoencoders to Clustering
Robert_AIZI · 2024-10-28T14:48:16.744Z · comments (0)
Ideas for Next-Generation Writing Platforms, using LLMs
ozziegooen · 2024-06-04T18:40:24.636Z · comments (4)
LessWrong email subscriptions?
Raemon · 2024-08-27T21:59:56.855Z · comments (6)
D&D.Sci Hypersphere Analysis Part 3: Beat it with Linear Algebra
aphyer · 2024-01-16T22:44:52.424Z · comments (1)
[question] Thoughts on Francois Chollet's belief that LLMs are far away from AGI?
O O (o-o) · 2024-06-14T06:32:48.170Z · answers+comments (17)
How do LLMs give truthful answers? A discussion of LLM vs. human reasoning, ensembles & parrots
Owain_Evans · 2024-03-28T02:34:21.799Z · comments (0)
The Sequences on YouTube
Neil (neil-warren) · 2024-01-07T01:44:39.663Z · comments (9)
Distillation of 'Do language models plan for future tokens'
TheManxLoiner · 2024-06-27T20:57:34.351Z · comments (2)
Optimizing Repeated Correlations
SatvikBeri · 2024-08-01T17:33:23.823Z · comments (1)
Causality is Everywhere
silentbob · 2024-02-13T13:44:49.952Z · comments (12)
Improving SAE's by Sqrt()-ing L1 & Removing Lowest Activating Features
Logan Riggs (elriggs) · 2024-03-15T16:30:00.744Z · comments (5)
Trying Bluesky
jefftk (jkaufman) · 2024-11-17T02:50:04.093Z · comments (16)
My Dating Heuristic
Declan Molony (declan-molony) · 2024-05-21T05:28:40.197Z · comments (4)
Exploring OpenAI's Latent Directions: Tests, Observations, and Poking Around
Johnny Lin (hijohnnylin) · 2024-01-31T06:01:27.969Z · comments (4)
[link] what becoming more secure did for me
Chipmonk · 2024-08-22T17:44:48.525Z · comments (5)
Sleeping on Stage
jefftk (jkaufman) · 2024-10-22T00:50:07.994Z · comments (3)
[link] Agreeing With Stalin in Ways That Exhibit Generally Rationalist Principles
Zack_M_Davis · 2024-03-02T22:05:49.553Z · comments (22)
AI #57: All the AI News That’s Fit to Print
Zvi · 2024-03-28T11:40:05.435Z · comments (14)
SAE features for refusal and sycophancy steering vectors
neverix · 2024-10-12T14:54:48.022Z · comments (4)
Consequentialism is a compass, not a judge
Neil (neil-warren) · 2024-04-13T10:47:44.980Z · comments (6)
← previous page (newer posts) · next page (older posts) →