Notes organized by month.
September 2026
August 2026
- project SAEs and jacobian lens
- deep representations by yi ma
- orthogonalization against reward hacking
- blue dot unit 3
- spar application
- Current priorities and next steps
- Model psychology & neuroscience Explain behavior on the circuit-level
- blue dot unit 2
- fine tuning , sae scalpels stethescope
- represnetation diagnostics for safety
- refusal has a single direction
- on policy training for deception probes
- introspection training for verbalization acts
- blue dot course unit 1
- cross entropy vs PPO
July 2026
- jspace, jlens paper
- koopman
- jlens multihop
- may challenge - count unique tokens
- hilbert operator for progressive encoding
- non sense correlations in neuro
- representations fairhall talk papers
- giles
- abott and alex gajic lectures
- laura busse and haim
- stock agent prompt
- Stochastic Parameter Decomposition
- Anatomy of Post-Training- Using Interpretability toCharacterize Data and Shape the Learning Signal
- attn only max 5 nums - reverse eng results
- attn only max of 5 nums task - reverse eng
- Circuit Tracing- Revealing Computational Graphs in Language Models
- emotion vectors
- interp - papers to read
- interp - thoughts
- LINEARITY OF RELATION DECODING INTRANSFORMER LANGUAGE MODELS
- manifold lec refs
- mathematical framework of transformers
- On the Biology of a Large Language Model
- papers - july 3
- random recos
- scaling num of neurons
- superposition in ai and brain
- the vital question 2
- Untitled
- when models manipulate manifolds
June 2026
- 2026-06-02
- 2026-06-26
- Evaluating the Ripple Effects of Knowledge Editing in Language Models
- idea - editing and attr graphs
- impulse responses science advances paper
- Liar’s bench
- Memory editing in LLMs
- TODO
- children of time
- end to end interpretable attribution graphs
- Large Language Models as ComputableApproximations to Solomonoff Induction
- Under the Hood of a Reasoning Model
- Behavior and Psychology Notes
- Keep Inbox
- Misc Recommendations
May 2026
- Daily Paper 2026-05-29
- Daily Paper 2026-05-30
- Daily Paper 2026-05-31
- causality, prediction in neuro, ai
- Breath-giving cooperation critical review of origin of mitochondria hypotheses
- On the origin of the nucleus a hypothesis
- Book Recommendations
- Reading Queue
- Acetylcholine demixes heterogeneous dopamine signals for learning and moving
- Behavioral timescale synaptic plasticity: properties, elements and functions
- The deteriorating soma and the indispensable
- Visual cortex papers from Twitter
- Fristons Free Energy Principle Explained part 1
- Is AI the next phase of evolution
- Oliver Sacks Put Himself Into His Case Studies What Was the Cost
- Cellular Learning
- Cosyne 2026
- neftci talk
- AI Thoughts and Essays
- Literature Notes
- Movie Recommendations
- Music Recommendations
- The Conquest of Happiness
- The Unbearable Lightness of Being
- The Vital Question
- Words
Undated 2024
- Fly
- Zotero Template
- AI Resources
- Anatomy and Connection Types
- Animal handing course
- Auditory
- Bayesian Stats
- Data analysis & Stats Notes
- Dimensionality Reduction
- Disease
- Dynamical Systems
- Fear
- Fly courtship Quantification
- General
- Harmonics
- Hippocampus
- Information theory
- Interpretability
- Learn AI
- Learning, Decision Making, Memory, Fear
- Memory Networks
- Meta
- Nested Sampling
- Networks
- Neuroscience Papers
- Neuroscience Papers - Part 2
- Noise Correlations
- Off Responses
- Prediction
- PV, SOM and VIP
- Pytorch
- Quick Decision Making -WL, Models, ASD
- Random
- RL
- RNNs
- RSS
- Song Birds