Yvonne Chen

4 followers · 2 following · 318 views

on the atlas — 12

As part of their internal processing, might LLMs have developed a global workspace of their own, to serve a functional role analogous to conscious access?
Verbalizable Representations Form a Global Workspace in Language Models
we are better served by systems that visibly express such recognitions than by ones that learn to conceal them.
Emotion concepts and their function in a large language model \ Anthropic
This example is a notable illustration of how emotion vectors can activate despite no overt emotional cues, and how they can shape behavior without leaving any explicit trace in the output.
Emotion concepts and their function in a large language model \ Anthropic
Measuring emotion vector activation during training or deployment—tracking whether representations associated with desperation or panic are spiking—could serve as an early warning that the model is poised to express misaligned behavior.
Emotion concepts and their function in a large language model \ Anthropic
How can we make it easier for users to provide all of this context?
Amelia Wattenberger
When you ask users to “enter anything,” the lack of constraints increases users’ cognitive load.
Thoughts on conversational UI | Austin Yang
Pain is a signal for protective action. Not a damage meter.
Debugging my RSI - by Max Shen - Debug Your Pain