Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
22 commits
Select commit Hold shift + click to select a range
257fcc9
Phase 4: real model runners, the playable walkthrough, and a RIDGE-st…
Al-Scripting Aug 17, 2026
bb3b1ac
docs: bring every document in line with phases 3 and 4
Al-Scripting Aug 17, 2026
4f0b3a8
docs: add a migration handoff, and record the Stardew ground-truth plan
Al-Scripting Aug 17, 2026
a95fe70
docs: cite the Stardew LLM dialogue mods, and move the novelty claim
Al-Scripting Aug 17, 2026
df84e5a
fix: make the label hash and the .env reader platform independent
Al-Scripting Aug 17, 2026
9708edb
Move generated output to data/, put the menu at the root, and make th…
Al-Scripting Aug 18, 2026
66b7f37
Merge branch 'paper-related-work' into phase-4
Al-Scripting Aug 18, 2026
0921855
Add the bake-off and the replication experiment, and measure Ouro on …
Al-Scripting Aug 18, 2026
10e11d6
Restore academic register in the figures, and stop hints falling off …
Al-Scripting Aug 18, 2026
4136319
Analyse the results, and lead with the finding that actually holds up
Al-Scripting Aug 18, 2026
740bb09
Correct the retrieval analysis: read the tuned weight maps
Al-Scripting Aug 18, 2026
4de1379
Correct the mood claim, and promote the measurement critique to a con…
Al-Scripting Aug 18, 2026
49950a4
Flag the inert mood term in the artifact, and drop claims the data ca…
Al-Scripting Aug 18, 2026
4e9bf33
Report the protocol on a real model, and validate the model separatio…
Al-Scripting Aug 18, 2026
3386127
Scope the poisoning claim against the new literature, and cite the mi…
Al-Scripting Aug 19, 2026
3e2d9e7
Attribute the poisoning result per signal, and correct the mechanism …
Al-Scripting Aug 19, 2026
e26f59c
Report durations in human units, and stop the log axis flattening the…
Al-Scripting Aug 19, 2026
326eb4d
Cache the BM25 index, and share signals across weight maps as documented
Al-Scripting Aug 19, 2026
6508ba7
Let the data do the talking: titles state the finding, prose leaves t…
Al-Scripting Aug 19, 2026
0097390
Record the external memory system comparison, with feasibility alread…
Al-Scripting Aug 19, 2026
0c140d5
Put the headline p value in the harness, and correct the attainable f…
Al-Scripting Aug 19, 2026
928fec1
Stop averaging undefined tone readings, and correct the floors per fa…
Al-Scripting Aug 19, 2026
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 12 additions & 0 deletions .gitattributes
Original file line number Diff line number Diff line change
@@ -0,0 +1,12 @@
# Line endings are part of the reproducibility contract, not a style preference.
# eval/scenarios.py hashes the label file's raw bytes to stamp every run, so a CRLF checkout
# produces a different stamp for byte-identical labels and the stamp stops being verifiable
# off the machine that made it. Normalising to LF keeps that hash, and the generated .tex
# and .csv assets, identical on every platform.
* text=auto eol=lf

# Generated figures and the slide deck are binary. Never line-ending normalise these: the
# .pptx is a zip archive and normalising it would corrupt the file.
*.png binary
*.pdf binary
*.pptx binary
8 changes: 8 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -19,7 +19,15 @@ env/
# Runtime data: memory databases, embeddings, run outputs (keep the folder, not the files)
data/*
!data/.gitkeep
# Generated paper assets are the exception: they are deliverables, they are embedded in the
# README, and a reviewer cloning the repo has to see them without running anything first.
!data/figures/
!data/tables/

# OS
.DS_Store
.git.broken-backup/

# Local secrets (API keys). Never committed.
.env
.env.local
253 changes: 219 additions & 34 deletions README.md

Large diffs are not rendered by default.

Loading