Skip to content

docs: G2a full-set 4.5701 — attribution stands #105

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
ronaldtse wants to merge 1 commit into from
Closed

docs: G2a full-set 4.5701 — attribution stands #105

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
13 changes: 13 additions & 0 deletions docs/RESULTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -423,3 +423,16 @@ The law sharpens: the pretrained WIDTH is load-bearing - projection
destroys the representation at any compression; depth is the
compressible axis. First verdict carrying the label-provenance hash
(labels sha256 e70ce991..., 137.2MB recorded in final_eval.json).

## ara-diac-small-2-6ep — full-set verdict: 4.5701; the discipline vindicated (2026-08-31)

Full 1,200-paragraph gate (teacher in-run 2.289): student **4.5701**,
gate_delta 2.28. Doubling epochs bought 0.25pp over the 3-epoch 2.0
rung (4.8218 -> 4.5701) - the residual is NOT mostly undertraining;
the E2/E3 domain attribution substantially stands (~0.25pp epochs,
~2.0pp domain/other). The subset had said 2.0062 / delta 0.72 - a
near-overturn of the decomposition that the full set corrects: the
FOURTH subset-overstatement instance and the most dramatic (subset
delta 0.72 -> full delta 2.28, a 3.2x inflation). The full-set-only
publication rule earns its keep here: this entry is the paper's
strongest measurement-discipline exhibit.