Topic
Jacobian Lens and Global Workspace Interpretability
A 2026 Transformer Circuits study maps intermediate activations into final-layer coordinates, then uses interventions to test a small token-aligned broadcast component.
Blog
Writing and Updates
Posts plus selected topic and lab updates that changed how a learner moves through the site.
Updates
Recent changes
Topic
Ring vs. Tree Attention
An interactive comparison of ring and balanced-tree schedules for combining distributed online-softmax states, with exact-arithmetic equivalence and scoped communication costs.
Topic
AlphaProof and AI-Assisted Theorem Proving
An updated guide to AlphaProof's three Lean-checked IMO 2024 solutions, AlphaGeometry 2, reinforcement-learning proof search, and the kernel trust boundary.
Topic
Model Comparison Table
A July 11 snapshot of GPT-5.6 Sol, Terra, and Luna; GPT-Live; Claude Fable 5 and Sonnet 5; restricted Mythos 5; and other major families, using official release material and leaving undisclosed architecture fields unknown.
Topic
FSRS: Spaced-Repetition Scheduling as Parameter Estimation
A statistical view of the difficulty-stability-retrievability model, its power-law forgetting curve, parameter fitting, and TheoremPath's fixed FSRS-5 defaults.
Lab
Diffusion Lab
Step the forward and reverse processes of a 2D toy diffusion. Watch noise schedules, score estimates, and DDIM vs DDPM samplers side by side.
Lab
Induction Heads Lab
Train a 2-layer attention-only transformer in your browser. Watch the induction circuit form during the loss-cliff phase transition; ablate any head to see in-context learning collapse.
Lab
High-Dimensional Probability Lab
Thin shells, random projections, Marchenko-Pastur spectra, and spiked PCA on one interactive board.
Writing
Posts
What OpenAI's goblin episode reveals about reward models
A small creature-metaphor habit in GPT-5.5 becomes a clean case study in reward models, proxy objectives, behavior transfer, and synthetic-data feedback loops.
ReadFree Energy Is a Serious Language for Generative AI
A book note on Welling, Lu, and Holdijk's Generative AI and Stochastic Thermodynamics: variational free energy as a unifying language for VAEs, normalizing flows, diffusion, flow matching, and Schrödinger bridges.
ReadScaling Made AI Legible
A book note on The Scaling Era: An Oral History of AI, 2019-2025. Scaling made AI progress legible. The next edge will come from research, judgment, and tool fluency.
ReadFast Tools, Slow Understanding
A short note on fast tools, fluent output, and the slower work of understanding machine learning.
Read