Skip to content

[TEST ONLY]Drop scaled_dot_product_attention #6367

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

Closed
wants to merge 1 commit into from
Closed

[TEST ONLY]Drop scaled_dot_product_attention #6367

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
6 changes: 5 additions & 1 deletion examples/models/llama/llama_transformer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,7 @@

# Please refer to README.md in the same folder for more information.

import math
from dataclasses import dataclass
from functools import partial
from typing import Dict, Optional, Tuple
Expand Down Expand Up @@ -251,7 +252,10 @@ def forward(

k = k.repeat_interleave(self.n_rep, dim=1)
v = v.repeat_interleave(self.n_rep, dim=1)
y = F.scaled_dot_product_attention(q, k, v, attn_mask=attn_mask, dropout_p=0.0)
scores = torch.matmul(q, k.transpose(2, 3)) / math.sqrt(self.head_dim)
scores = scores + attn_mask
scores = F.softmax(scores.float(), dim=-1).type_as(q)
y = torch.matmul(scores, v)

return y.transpose(1, 2).contiguous().view(bsz, seqlen, self.dim)

Expand Down
Loading