https://www.youtube.com/live/a6gj7E4AblI?si=xw534iWSpp7ZSwBD
тут критику підвезли, але поки у мене нема вільних 3.5 годин

https://x.com/vincentweisser/status/1867719020444889118


@ilyasut
full talk at neurips 2024 "pre-training as we know it will end" and what comes next is superintelligence: agentic, reasons, understands and is self aware

Optimized einsum https://optimized-einsum.readthedocs.io/en/stable/

ДЮНА

The Geometry of Concepts: Sparse Autoencoder Feature Structure

Abstract:

Sparse autoencoders have recently produced dictionaries of high-dimensional vectors corresponding to the universe of concepts represented by large language models. We find that this concept universe has interesting structure at three levels: 1) The "atomic" small-scale structure contains "crystals" whose faces are parallelograms or trapezoids, generalizing well-known examples such as (man-woman-king-queen). We find that the quality of such parallelograms and associated function vectors improves greatly when projecting out global distractor directions such as word length, which is efficiently done with linear discriminant analysis. 2) The "brain" intermediate-scale structure has significant spatial modularity; for example, math and code features form a "lobe" akin to functional lobes seen in neural fMRI images. We quantify the spatial locality of these lobes with multiple metrics and find that clusters of co-occurring features, at coarse enough scale, also cluster together spatially far more than one would expect if feature geometry were random. 3) The "galaxy" scale large-scale structure of the feature point cloud is not isotropic, but instead has a power law of eigenvalues with steepest slope in middle layers. We also quantify how the clustering entropy depends on the layer.

https://arxiv.org/abs/2410.19750