Skip to content
conftrace_

convergence dynamics

2 papers

Explore in graph

Co-occurring keywords · papers mentioning both

pattern identification (1) formula discovery (1) mathematical constant (1) continued fraction (1) least square (1) autoregressive training (1) self-attention model (1) generative model (1) gradient descent (1) in-context learning (1)

Papers

On Mesa-Optimization in Autoregressively Trained Transformers: Emergence and Capability NeurIPS 2024 Unsupervised Discovery of Formulas for Mathematical Constants NeurIPS 2024