Initial K3 snapshot: 0.5B KDA/MLA/MoE train path

Standalone tree split from LLMRL/projects/kda. Includes Triton dt_bias
backward fix, train_k3 --preset 0.5b, SFT, Docker runtime, and tests.
This commit is contained in:
dela
2026-08-25 14:43:17 +08:00
commit 584f7e9e73
140 changed files with 21592 additions and 0 deletions
+14
View File
@@ -0,0 +1,14 @@
from .cumsum import (
chunk_local_cumsum,
chunk_local_cumsum_scalar,
chunk_local_cumsum_vector,
)
from .index import prepare_chunk_indices, prepare_chunk_offsets
__all__ = [
"chunk_local_cumsum",
"chunk_local_cumsum_scalar",
"chunk_local_cumsum_vector",
"prepare_chunk_indices",
"prepare_chunk_offsets",
]