x

LESSWRONG

LW

hillz — LessWrong

hillz

hillz

Message

Research scientist & engineer, AI Safety thinker.

Machine-learner, meat-learner, model trainer, skeptical adorer of statistics.

Reach out if you'd interested in hiring me to join your alignment research team.

- Hillary Sanders

78

3

57

10y

hillz

Research scientist & engineer, AI Safety thinker.

Machine-learner, meat-learner, model trainer, skeptical adorer of statistics.

Reach out if you'd interested in hiring me to join your alignment research team.

- Hillary Sanders

Frontier AI Shops Should Be Filtering and Generating AI Discourse During Pretraining

We should be filtering out a huge amount of AI safety discourse, general AI discourse, and adversarial AI stories from LLM pretraining. We should also be seeding pretraining with generated stories of AIs helping and protecting humans. Why? Persona effects We know that LLMs are heavily influenced by the personas...

hillz's Shortform

Sparse Efficiency vs. Superposition: The Interpretability Tradeoff

Today’s frontier models train in an expensive style: dense forward passes, huge matrix multiplies, and broad weight updates. The human brain (~5 MWh over 28 years) is an existence proof that learning can be vastly more energy efficient - about 10,000x - than modern AI training runs (https://coefficientgiving.org/research/how-much-computational-power-does-it-take-to-match-the-human-brain/). The human...