REVIEW 4 cited by
Birth of a Transformer: A Memory Viewpoint
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Birth of a Transformer: A Memory Viewpoint
read the original abstract
Large language models based on transformers have achieved great empirical successes. However, as they are deployed more widely, there is a growing need to better understand their internal mechanisms in order to make them more reliable. These models appear to store vast amounts of knowledge from their training data, and to adapt quickly to new information provided in their context or prompt. We study how transformers balance these two types of knowledge by considering a synthetic setup where tokens are generated from either global or context-specific bigram distributions. By a careful empirical analysis of the training process on a simplified two-layer transformer, we illustrate the fast learning of global bigrams and the slower development of an "induction head" mechanism for the in-context bigrams. We highlight the role of weight matrices as associative memories, provide theoretical insights on how gradients enable their learning during training, and study the role of data-distributional properties.
Forward citations
Cited by 4 Pith papers
-
Fingerprint, Not Blueprint: How Positional Schemes Set the Default Spectral Algebra of Attention
Positional schemes set the default spectral algebra of attention heads: previous-token heads are rotational under RoPE and content-like under absolute/ALiBi, as a post-function fingerprint rather than a hard constraint.
-
Structure Before Collapse: Transient semantic geometry in next-token prediction
Semantic geometry emerges transiently early in next-token prediction training before collapsing to Neural Collapse symmetry in synthetic settings with latent semantic factors.
-
Sharp Capacity Thresholds in Linear Associative Memory: From Winner-Take-All to Listwise Retrieval
Winner-take-all linear memory capacity scales as d² ~ n log n due to extreme values; listwise retrieval via Tail-Average Margin yields d² ~ n with exact asymptotic theory.
-
Emergence of Frontier Superposition: M\"obius attractor and Cascade Supervision
Gradient descent reaches equal-weight superposition on Erdős-Rényi reachability via a Möbius attractor under S_n symmetry and Cascade Supervision, with a parameter-free decay law matching experimental cosines within 0.02.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.