Created: April 07, 2022
Modified: April 07, 2022
Modified: April 07, 2022
emergent capabilities
This page is from my personal notes, and has not been specifically reviewed for public consumption. It might be incomplete, wrong, outdated, or stupid. Caveat lector.A consequence of phase transitions in large models is that models may end up having capabilities we didn't expect. For example, models that are not explicitly set up as agents (e.g., language models) may eventually end up doing agentic reasoning internally.
Inducing such capabilities is debatably the entire point of meta learning: rather than program capabilities into the system directly, we set up conditions that encourage them to emerge.