Skip to content

fix: update datatype to int #150

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
kmehant merged 1 commit into from
Sep 16, 2025
Merged

fix: update datatype to int #150

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
2 changes: 1 addition & 1 deletion ...ins/fused-ops-and-kernels/src/fms_acceleration_foak/kernels/unsloth/cross_entropy_loss.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,7 +187,7 @@ def _cross_entropy_backward(
pass


MAX_FUSED_SIZE = tl.constexpr(65536) # 2**16
MAX_FUSED_SIZE = 65536 # 2**16

class Fast_CrossEntropyLoss(torch.autograd.Function):
@staticmethod
Expand Down
3 changes: 2 additions & 1 deletion plugins/fused-ops-and-kernels/src/fms_acceleration_foak/kernels/unsloth/rope_embedding.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@
import torch
from .utils import calculate_settings

ROPE_GROUP_SIZE = tl.constexpr(4)
ROPE_GROUP_SIZE : int = 4

@triton.heuristics({"BACKWARD_PASS": lambda args: args["BACKWARD_PASS"],})
@triton.jit
Expand All @@ -36,6 +36,7 @@ def _rope_embedding(
RoPE is Q * cos + rotate_half(Q) * sin
See our blog post for more info
"""
ROPE_GROUP_SIZE = 4
row_position = tl.program_id(0)
group_head_position = tl.program_id(1)
col_offsets = tl.arange(0, BLOCK_SIZE)
Expand Down