Skip to content

docs: propagate the 2.0 correction #191

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

Merged
ronaldtse merged 1 commit into from
Sep 5, 2026
Merged

docs: propagate the 2.0 correction #191

Show file tree
Hide file tree
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
1 change: 1 addition & 0 deletions docs/RESULTS.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -217,6 +217,7 @@ The viable path to a sub-100MB browser tier is width reduction FROM a
pretrained ByT5-small (closed-form stitch across widths, microkimi
protocol), not from-scratch training.

[CORRECTED 2026-09-05: the 4.8218 figure below did not reproduce; the corrected 2.0 number is 5.08 (see the correction entry above).]
## ara-diac-small-1.0 — Arabic client tier (2026-08-24)

Sequence-level KD from the r6 teacher (rababa_arabic_byt5/run-006-morph,
Expand Down
6 changes: 4 additions & 2 deletions models/ara-diac-small/ara-diac-small-2.0.README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,10 +6,12 @@ architecture and artifact size as 1.0: the **r7 canonical teacher**
(2.2864; fresh greedy labels) and the **Muon optimizer** (E3-adopted).

- 1.0 (r6 labels, AdamW): 8.26 full-set windowed DER-CE
- 2.0 (r7 labels, Muon): **4.82** (teacher r7 in-run: 2.289)
- 2.0 (r7 labels, Muon): **5.08** (teacher r7 in-run: 2.289; corrected
2026-09-05 — the first-published 4.82 did not reproduce, see
docs/RESULTS.md)

A 42% error reduction, pre-registered as E4 (gate ≤ 6.26; prediction
4.3–5.0 — landed at 4.82). The strict teacher+0.5pp gate is still
4.3–5.0 — landed at 5.08 after the 2026-09-05 correction). The strict teacher+0.5pp gate is still
missed (+2.53pp, disclosed); the E2/E3 factorial attributes the
residual to domain coverage. Identical IMF v1 contract: dynamic fetch,
sha256-verified, KV decode, margins JSON alongside.
Expand Down
4 changes: 2 additions & 2 deletions models/ara-diac-small/ara-diac-small-2.0.metadata.yaml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ trained_from: 'sequence-level KD from the r7 canonical teacher (rababa_arabic_by
2.2864 windowed DER-CE full protocol): fresh greedy r7 labels on the same r5-units
corpus/limits as ara-diac-small-1.0, Muon optimizer (E3-adopted), vanilla ByT5-small
(E4, pre-registered gate <= 6.26). Checkpoint rababa-checkpoints:/rababa_arabic_distill_small/run-006-r7-muon/best.
The two measured wins compound: 8.259 -> 4.8218 full-set windowed DER-CE (teacher
The two measured wins compound: 8.259 -> 5.0821 full-set windowed DER-CE (corrected 2026-09-05 (teacher
reproduces 2.289 in-run vs documented 2.2864) — a 42% error reduction on the 1.0
release at the same architecture and artifact size. Still misses the strict teacher+0.5pp
gate (+2.53pp; miss disclosed); the E2/E3 factorial attributes the residual to domain
Expand All @@ -25,7 +25,7 @@ metrics:
the documented 2.2864 (r7 canonical teacher)
source: interscript/interscript-ml docs/RESULTS.md#run-006-r7-muon
- name: der_student_fullset
value: 4.8218
value: 5.0821
protocol: same full-set harness; E4 (r7 teacher labels + Muon, vanilla ByT5-small)
vs the 8.259 AdamW/r6-labels 1.0 release — a 42% error reduction at identical
architecture and artifact size
Expand Down
Loading