Search Results for author:
Found 4 papers, 2 papers with code
Copy Suppression: Comprehensively Understanding an Attention Head
We show that self-repair is implemented by several mechanisms, one of which is copy suppression, which explains 39% of the behavior in a narrow task.
The Hydra Effect: Emergent Self-repair in Language Model Computations
We investigate the internal structure of language model computations using causal analysis and demonstrate two motifs: (1) a form of adaptive computation where ablations of one attention layer of a language model cause another layer to compensate (which we term the Hydra effect) and (2) a counterbalancing function of late MLP layers that act to downregulate the maximum-likelihood token.
Tracr: Compiled Transformers as a Laboratory for Interpretability
Additionally, the known structure of Tracr-compiled models can serve as ground-truth for evaluating interpretability methods.
Acquisition of Chess Knowledge in AlphaZero
In this work we provide evidence that human knowledge is acquired by the AlphaZero neural network as it trains on the game of chess.
