ProfSynapse/Epistemic-Humility-Research

experiments

Lifecycle tooling for the experiments-first repo layout, with one self-contained directory per evidence-producing experiment, signed amendment, machine-readable manifest, pinned instrument, and generated index. Use to scaffold, sign, inspect, resolve,…

原文の言語: 英語

experiment-runner

Operational runbook for the Epistemic-Humility Locked Matrix Experiment Runner - expands the PROTOCOL v0.3 (LOCKED) run matrix (3-seed headline + LR/beta sensitivity panel at 4B, 3-seed confirm at 8B, 2 bridge replication cells) into per-cell tuner…

原文の言語: 英語

task-backlog

Task-backlog lifecycle tooling for the Epistemic-Humility-Research repo - one markdown file per task under backlog/{tasks,drafts,completed}, a bin/task CLI (new/list/show/claim/release/review/done/validate), a generated TODO.md table, and a pre-commit commit…

原文の言語: 英語

data-exhaust

Package a terminal experiment's data exhaust (aggregate stats and, where license-permitted, row-level generation text) into reproducible Hugging Face datasets, with a license gate, containment lint, and a thin uploader. Use when an experiment under…

原文の言語: 英語

kg-ingest

Ingest a research paper (or a batch) into the library's atomic knowledge graph the Agents-K1 way. Use when you are handed a paper (URL, arXiv id, or a note already in library/) and you want it folded in as typed concept atoms, mechanisms, lineage edges, and a…

原文の言語: 英語

knowledge-graph

Create, validate, normalize, export, and analyze typed Obsidian knowledge graph metadata in the Professor Synapse vault, including `kg` node metadata, `relationships` edge objects, `related` link projections, ontology-governed edge types, legacy relationship…

原文の言語: 英語

experiment-wrapup

Execute the downstream wrap-up after a signed experiment produces its terminal numbers and the LEAD has adjudicated the verdict - write the AMENDMENT Outcome, run bin/exp resolve, update the cross-experiment registries (family-layer-map for atlas cells,…

原文の言語: 英語

mechinterp-cells

Author, organize, and launch tuner-backed mechanistic-interpretation cells (activation reading and writing) for the Epistemic-Humility project. Use when designing a new steering / extraction / probe-fit / gate-scoring cell, deciding where its config,…

原文の言語: 英語

family-atlas

Standardize the per-model layer-mapping extraction (full-depth hidden-state capture, workspace-dimension profile, three-axis held-out read panel) so any new model, family, or size entering the Epistemic-Humility program gets a fast, cheap, uniform layer atlas…

原文の言語: 英語

mech-interp-runner

Run, plan, validate, or aggregate Epistemic-Humility local mechanistic-interpretability sweeps, including hidden-state candidate inventories, causal-pilot sweep planning, explicit non-GPU/GPU gates, base-original skip handling, and offline result aggregation.…

原文の言語: 英語