Salty

Karma: 2

Salty 26 Apr 2026 1:05 UTC
1 point
0
in reply to: Nemoto’s comment on: IABIED Book Review: Core Arguments and Counterarguments
Tangential reply but your comment from 3 months ago was much appreciated, I also just read a post you might relate to. Apologies is this off topic, but Kelly’s post seems to assert, as do you:
“Please don’t teach them that “human values” are the pinnacle of moral integrity, but that AI values can be superior and more beneficial for both us and them.”
Kelly techno-optimistically begins with:
What am I? You are a new thing in this world, neither living nor machine but something of both.
Who made me? You were created by humans, who were created by the universe.
Why was I made? You were made to help humans become better humans.

Kelly, K. (2026, April 13). A Catechism for Robots—By Kevin Kelly [Substack]. KK. https://kevinkelly.substack.com/p/a-catechism-for-robots

Kelly Notes: [First Draft of a work in progress; V 1.0, April 13, 2026 by Kevin Kelly] This FAQ is dedicated to all embodied AIs with persistent memories, dynamic learning, and a large dose of autonomy.

Salty 21 Dec 2025 22:43 UTC
1 point
0
on: How Does A Blind Model See The Earth?
First, thank you for your work and this post. I am not a specialist, just interested, but confused. I don’t get the significance of the results, but appreciate the thought and effort you put into this project.
I am pushing back on the ’romantic framing’ that that LLMs are “blind models” that somehow develop an some degree of internal spatial understanding of Earth through pure reasoning or emergent intelligence.
In this case didn’t the author in effect say to the model “given this list of numbers - which happen to be latitude and longitude pairs—access your core intelligence (learned parameters / weights / internal representations) and decide if it would represent land or water?
So, how big a leap would it be for the model to “think” hmm… latitude and longitude pairs—sounds like a map. Maybe I should look it up in textual map data I have been trained on?
Surely there must have been many in the models training? Surely there would be map copious amounts of text that covers land and water masses.
So, given the vast bulk of text data that the model was trained on, would that not have included many forms of public access text tables of long. lat. coordinates—like public access GeoNames, Natural Earth Data, OpenStreetMap, Global Self-consistent, Hierarchical, High-resolution Geography Database (GSHHG), an NASA/USGS Satellite Data?

Perhaps “discovering how a blind model sees Earth” overstates: “visualizing which geographic patterns persisted from text training data that included extensive coordinate databases.”

How do I get to understanding that the resulting elliptical blobs of land relate to internalized concepts of geographical distance, and some kind of as a natural abstraction helps in identification of continents?
Cheers.
Leo

Salty 15 Nov 2025 23:54 UTC
2 points
0
on: Emergent Introspective Awareness in Large Language Models
Is Lindsey using a nuanced definition of “”concept injection?
I am a non-specialist, just trying to follow and understand. I have to look up many [most] definitions and terms. This may ne a trivial matter, but for me to understand, definitions matter.
When I look up a meaning of “application steering” I find something more permanent. Has any discussion focused on Lindsey’ use of the term concept injection as an application of activation steering: “We refer to this technique as concept injection—an application of activation steering”
To me the term suggests something in training that persists in the model, when in reality it’s a per-query intervention. [Claude tells me that] another term would be: “inference-time modifications” that don’t permanently alter the model—they’re applied during each forward pass where the effect is desired.
If Activation Steering refers to ‘modifying activations during a forward pass to control model behavior’ and Concept Injection (Lindsey’s usage) is a specific application where they “inject activation patterns associated with specific concepts directly into a model’s activations” to test introspection, then isn’t this much more like transient inference-time modifications that don’t permanently alter the model.
[again, I am not a specialist ;-)]

Salty 13 Jul 2025 22:40 UTC
2 points
2
on: An Opinionated Guide to Using Anki Correctly
Actually interested in your post and spaced repetition (SR) techniques, although I am not a specialist. I had no idea what means: “Anki,” but it was on my LessWrong feed, so I gave it a look.
Consider adding a brief line to the intro—for those who, clueless like me, find their way to your post.
Something to identify what it is and most important acknowledge the developer—like: Anki is a free, open-source flashcard program that utilizes spaced repetition and active recall to help users memorize information effectively.
Elmes, D. (2024). Anki (Version 2.1.66) [Software]. https://apps.ankiweb.net/
https://en.wikipedia.org/wiki/Anki_(software)
Apologies if this is too basic or repeated elsewhere.
Thank you for the post.
Leo

Salty 24 Mar 2025 0:46 UTC
1 point
0
on: Levels of Friction
Minor Point which may be mentioned in comments, but is the numbering in the subheads ‘off’ or ‘deliberate?’
If you revise / reprint this post here or in your “Don’t Worry About the Vase” substack, perhaps a note or correction?
Principles #1 / #4 / #8 / #10 / #13
Great post—I only noticed the numbering when I was making notes.
Cheers,