19 - Mechanistic Interpretability with Neel Nanda
AXRP - the AI X-risk Research Podcast
00:00
Language Model Interpretability
In language models it's often been easier to understand things at the output than things near the inputs. Unlike classic neural networks or convolutional networks where the key thing is you've got layers of neurons each layer's input is the output of the previous layer essentially. In standard framings of things like residual image networks this is often framed as a skip connection around the layer that's like a random thing you add on but isn't super important. The way transformers seem to work in practice is that every layer is kind of incrementally updating the central residual stream.
Transcript
Play full episode
