Bayes Blast 13 – GPT-4 maps every neuron in GPT-2

The Bayesian Conspiracy - A podcast by The Bayesian Conspiracy

Categories:

Blast with Matt re OpenAI using GPT-4 to automatically write explanations for the behavior of neurons in GPT-2, and what this means for Doom (spoiler: it’s good!) OpenAI’s explanation Thread – less than 1% of neuron’s explained with good confidence