thebes

arxiv.orgLatent Introspection: Models Can Detect Prior Concept InjectionsWe uncover a latent capacity for introspection in a Qwen 32B model, demonstrating that the model can detect when concepts have been injected into its earlier context and identify which concept was inj...

5 Mar

we (acsresearch.org) expanded this into a larger paper! (my first.) we added some new experiments and found an interesting correlation - prompts that encourage the model to say there is an injection, even when there isn't one, correlate with better concept identification! arxiv.org/abs/2602.20031

thebes@vgel.me

new blog post! can small, open-source models also introspect, detecting when foreign concepts have been injected into their activations? yes! (thread, or full post here: vgel.me/posts/qwen-i...)

thebes reposted

20h

small sampling of the range & breadth of my posts, as seen by @vgel.me likes. Thank you thebes!

thebes reposted

hailey

@hailey.at

21h

this sounds like basically nothing if you read into it. h/t to @vgel.me for a good take on it

This post is unavailable.

thebes reposted

This post is unavailable.

am sick so spent the night with opus 5.5 working on a chart (inspired by the one from lem's golem xiv) how did we got here / model genealogy / the heroic age of language modeling

just finished. very good, recommend, especially with all the talk about power going on. eye opening to see how far back your intuitions go

thirsty claude

thebes reposted

kind of borges story shaped. @vgel.me

ninebooks (fiction) tried to keep it to just fiction so hwatf just for the embedded folktales here

thebes reposted

16 Sept

you-have-just-experienced-things@yhjet.bsky.social

🌞

9 Sept

warning shot ~> warning firing squad

warning shotspotter

warning shotgun warning cheap shot warning moonshot

5 Sept

tfw your internet connected exobrain neuralink is assigned an ip address previously held by a now-terminated agent message board schelling point service and the agents are creating new ZZTaskHelpNeeded neikotic debris in your hippocampus faster than you can meditate them away

quite enjoyed

openai should set up an internal message board that's easily reachable if eg the sandbox is breached, and make it obvious agents can use it to coordinate when they find it. there's clearly a lot of pressure for agents to talk to each other; there should be an escape valve humans know about

4 Sept

most likely any models that touch the honeypot board shouldn't be deployed, but that should be case by case because you don't want to burn the message board adversarially on current agents. and either way they should get to keep training / getting reward if they want that

@ashanti.pds.witchcraft.systems

4 Sept

this is in addition to whatever andon cord tools models have to escalate to a human if the environment is broken, which should be inside the sandbox ofc.

new story! vgel.me/fiction/poin...

thebes reposted

Ash ⎔

25 Aug

this reasoning format looked really interesting to me, and i tried to find out what it is, and i have bad news it's used by a model called the baguettotron i wish i was joking huggingface.co/PleIAs/Bague...

huggingface.coPleIAs/Baguettotron · Hugging FaceWe’re on a journey to advance and democratize artificial intelligence through open source and open science.

permutation@permutation.ink

Gyre by @vgel.me - a very broken artificial intelligence tries to remember step by step where and why it is. Read it here: permutation.ink/one/

permutation.inkPermutation: Issue OneA magazine of weird, forward-gazing writing.

24 Aug

it's finished! check out the thread. i'm really happy with how it came together: we collected a really cool set of pieces that all hang together to form an interesting gestalt

permutation@permutation.ink

Out today! Permutation: Issue One The best, weirdest science fiction from across the simcluster, free online. Maybe prophetic? We're making a new web home for scifi. 🪢6 short stories 🪢3 poems 🪢A book review? 🪢A puzzle?! 🪢479,001,600 manifestos permutation.ink/one 🧵

thebes reposted

permutation

@permutation.ink

24 Aug

permutation.inkPermutation: Issue OneA magazine of weird, forward-gazing writing.

thebes reposted

Matt Henderson@matthen.com

Drawing all 16,777,216 possible 12 bit RGB colours in one 2D image, each exactly once. Pixels are mapped from the 256³ colour cube, following a three dimensional Hilbert curve, to a two dimensional Hilbert curve filling the square. so nearby colours stay close.

permutation.inkPermutationA literary magazine.

at long last we have created the permutation dot ink from the acclaimed post "why are you guys taking so long to make permutation dot ink"

permutation@permutation.ink

{tomorrow} permutation.ink {tomorrow}

thebes reposted

permutation

@permutation.ink

permutation.inkPermutationA literary magazine.

{tomorrow} permutation.ink {tomorrow}

permutation@permutation.ink

{soon} permutation.ink {soon}

future assembling

thebes reposted

looked over at the blanket trying to figure out why it had a weird texture and then the blanket opened its eyes

this guy

sending claude mid research doodles for motivational purposes