Loading: convert transposed and RoPE-permuted weights before taking the DTensor shard #48519
Load: convert transposed and RoPE-permuted weights in full before tak…
bb3a775d
Let conversion ops declare the dims they move; defer the shard only w…
9e2bca95
Keep the decision in one helper
d83a1f72
vasqu
commented
on 2026-09-04
Merge branch 'main' into load-convert-before-shard
324b3116
Narrow to the RoPE permutation, leave transposes to shard-on-read
f4562cf1
Merge branch 'main' into load-convert-before-shard
d8c8613a
Merge branch 'main' into load-convert-before-shard
7d960108
Merge branch 'main' into load-convert-before-shard
717a3a95
Merge branch 'main' into load-convert-before-shard
5892c331
Only build the sharding op when it is used
cb61b1aa
Let each op say whether it breaks shard-on-read
071db775
Merge branch 'main' into load-convert-before-shard
5fa9a03d
Merge branch 'main' into load-convert-before-shard
10f60888
Merge branch 'main' into load-convert-before-shard
14fe0333
Merge branch 'main' into load-convert-before-shard
65b6e406
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub