Conscious Data
Search
Search
Search
Dark mode
Light mode
Explorer
tags
groups
selections
links
"LLMs work best by simulating engaged people"
a chat about a FUSE filesystem which lazily renders nested contents with LLMs
Cyborgism (post)
decoupling protocol
fermi location
hella-bangers
Hermes Prompt
instantiating a Sydney Waluigi, 5 March 2023
Janus' GPT Wrangling
Janus' Simulators
janus's favorite book
LessWrong
Loom – interface to the multiverse
most entities fund people, not ideas
notes on Karpathy's Intelligence Amplification
on agency, in response to Michael Nielsen
on AI as a science
on attempts to commodify the act of prompting a language model
on insights to be found in mechanistic+developmental interpretability
on the importance of setting strong principles
on the prevalence of a deep semantic skepticism
on truesight and operation of base models
roon on industrial understanding of model behavior
roon's @repligate research program
Simulacra and simulation (from "Role play with large language models", Nature)
Simulacra are Things
Simulators
Simulators reflection, Dec. 2023
Simulators sequence
the essence of making yourself clear to a machine is to think clearly yourself
thoughts on the Waluigi Effect, Feb. 2023
what if I auto-corrected you
Why I take short timelines seriously
wisdom hypothesis
You won't measure what you don't understand
LLMs
Loom
models
Bing
Claude
Whisper
people
ampdot
anthrupad
celeste
deepfates
Janus
John David Pressman
Katan'Hya
Linus Lee
roon
Scott Alexander
tel∅s
platforms
are.na
discord
obsidian
projects
catalog
colorweave
contextualize
mt-img
promptlib
agency
AI Alignment
Media
prompting
researchers
screenshots
Home
❯
tags
❯
groups
❯
selections
selections
Jun 12, 2024, 1 min read
group:
e4a9d672-2846-4bc5-9107-8761bc5b7ba8
Graph View
Backlinks
No backlinks found