Skip to content

Commit aef91ea

Browse files
author
Ronald Tse
committed
feat(export): ara-diac-small-2.0 — the Muon optimizer rung (8.26 -> 5.29)
Same corpus/labels/epochs as 1.0, optimizer as the only variable (E3 factorial vanilla arm). Full-set DER disclosed against the strict gate; r7 re-distillation noted as the expected successor.
1 parent e1e7459 commit aef91ea

3 files changed

Lines changed: 61 additions & 0 deletions

File tree

Lines changed: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,20 @@
1+
# ara-diac-small-2.0
2+
3+
Arabic diacritization (haraqat restoration). Client-tier ByT5-small
4+
student, identical corpus/labels/epochs to ara-diac-small-1.0 with one
5+
variable changed: **the Muon optimizer** (E3 factorial). That single
6+
change closes 2.96pp of the 5.68pp teacher-student gap:
7+
8+
- 1.0 (AdamW): 8.26 full-set windowed DER-CE
9+
- **2.0 (Muon): 5.29** (teacher r6: 2.58; in-run reproduction 2.60)
10+
11+
The strict teacher+0.5pp gate is still missed; the residual decomposes
12+
as ~0.70pp capacity + ~2.25pp domain coverage (E2/E3 factorial), and an
13+
r7-teacher re-distillation is in flight. Identical IMF v1 contract:
14+
dynamic fetch, sha256-verified, KV decode, margins JSON alongside.
15+
16+
```python
17+
from interscript_ml import Model
18+
model = Model.load("ara-diac-small-2.0")
19+
model.translate("قوله فحكمها في الوفاة")
20+
```
Lines changed: 32 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,32 @@
1+
format: imf-v1
2+
id: ara-diac-small-2.0
3+
task: diacritization
4+
source_script: Arab
5+
target: Arab
6+
tokenizer: bytes
7+
opset: 14
8+
decoder: kv
9+
precision: fp32
10+
license: BSD-3-Clause
11+
trained_from: 'sequence-level KD from the r6 teacher (rababa_arabic_byt5/run-006-morph/best,
12+
2.5793 windowed DER-CE full protocol): the same 29,322 greedy teacher labels as
13+
ara-diac-small-1.0, with the Muon optimizer (E3 factorial, vanilla student arm).
14+
Checkpoint rababa-checkpoints:/rababa_arabic_distill_small/run-005-muon/best. The
15+
optimizer alone closes 2.96pp of the 5.68pp teacher-student gap measured for the
16+
1.0 (AdamW) release: 8.259 -> 5.2945 full-set windowed DER-CE (teacher reproduces
17+
2.5997 in-run). Still misses the strict teacher+0.5pp gate — the residual decomposes
18+
as ~0.70pp capacity + ~2.25pp domain coverage per the E2/E3 factorial; miss disclosed.
19+
An r7-teacher re-distillation is in flight and expected to supersede as a later rung.'
20+
metrics:
21+
- name: der_teacher_fullset
22+
value: 2.5997
23+
protocol: windowed DER-CE (1400-byte windows, word-boundary split, greedy,
24+
haraqat-projected, Misraj evaluator); full 1,200-paragraph SadeedDiac-25;
25+
in-run reproduction of the documented 2.5815/2.5793
26+
source: interscript/interscript-ml docs/RESULTS.md#run-005-muon
27+
- name: der_student_fullset
28+
value: 5.2945
29+
protocol: same full-set harness; +2.69pp over the in-run teacher — the Muon
30+
arm of the E3 2x2 factorial (vanilla student, optimizer as the only variable
31+
vs the 8.259 AdamW release)
32+
source: interscript/interscript-ml docs/RESULTS.md#run-005-muon

‎src/gpu/modal_export.py‎

Lines changed: 9 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -114,6 +114,15 @@
114114
"test_data": "arabic-sadeed-imf/test.jsonl",
115115
"probe": "قوله",
116116
},
117+
"ara-diac-small-2": {
118+
"volume": "/volumes/rababa-checkpoints",
119+
"checkpoint": "rababa_arabic_distill_small/run-005-muon/best",
120+
"metadata": "models/ara-diac-small/ara-diac-small-2.0.metadata.yaml",
121+
"readme": "models/ara-diac-small/ara-diac-small-2.0.README.md",
122+
"test_volume": "/datasets/rababa",
123+
"test_data": "arabic-sadeed-imf/test.jsonl",
124+
"probe": "قوله",
125+
},
117126
"urd-diac": {
118127
"volume": "/volumes/urdu-diacrit-checkpoints",
119128
"checkpoint": "urdu_diacrit/run-001/best",

0 commit comments

Comments
 (0)