The upper bound of what can be learned from a dataset is not the most capable trajectory, but the conditional structure of the universe implicated by their sum.

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Unsupervised learning builds a low-dimensional model of the input data.claim0.789
  Clarifies what unsupervised learning does.
• Information from point A to B can travel through C(m+n, n) distinct paths, which quickly exceeds the number of atoms in the visible universe.claim0.780
  Janus's mathematical claim about exponential path combinatorics in transformers.
• Learned simulations can be partially observed and lazily-rendered, and still work.claim0.767
  One of the updates about prosaic ML simulation.
• If loss keeps going down on the test set, in the limit the model must be learning to interpret and predict all patterns represented in language, including common-sense reasoning, goal-directed optimization, and deployment of the sum of recorded human knowledge.hypothesis0.766
  Extrapolation of scaling predictive models to AGI.
• When a model discovers that its outputs produce effects, it accelerates learning through in-context learning, analogous to lucid dreaming.claim0.765
  Describes scaffolding method and the model's meta-learning loop.
• Curiosity, insight, decision-making, and diverse phenomena can all be accommodated by a single imperative: minimization of expected free energy (resolution of uncertainty).claim0.765
  Central thesis of the paper unifying cognitive phenomena under one objective function
• Temporal discounting emerges naturally from active inference without an explicit discount factor, because predictions in the distant future are less precise.claim0.759
  §2, discussion of precision.
• The two-dimensional subspace reported by Burger et al. (2024) seems to reflect a stage of transition in the model's processing, rather than a universal property of truth directions.quote0.759
  Load-bearing interpretive claim about the layer-specificity of Burger et al.'s finding.