1 citations · 1 across the 5 of their papers we have counts for
3 papers · 1 filter
In-Context Learning Before Specialisation: Theory of Learning in a Solvable Topic Model
Francesco Cagnetta, Casper L. Christensen
Sequence models can generalise in fundamentally distinct ways: extrapolating structures seen in training, or using their prompt to adapt to novel tasks. The prevailing strategy det…
LLMs program using Reusable and Composable Abstract Representations
Casper L. Christensen, Kyle J. Ray, Dmitry Manning-Coe +2
When writing code, the same computation can be expressed in different programming languages and through different implementation strategies. We show that the activations of LLMs ca…
Transformers learn factored representations
Adam Shai, Loren Amdahl-Culleton, Casper L. Christensen +6
Transformers pretrained via next token prediction learn to factor their world into parts, representing these factors in orthogonal subspaces of the residual stream. We formalize tw…