Skip to content
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

Jump to bottom

use fast softmax only on prefill #1159

Merged
merged 2 commits into from
Aug 5, 2024
Merged

use fast softmax only on prefill #1159

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b3c450a
use fast softmax only on prefill (#306)
jaygala223 Jul 17, 2024
d338f55
make style
ssarkar2 Aug 5, 2024
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 optimum/habana/transformers/models/llama/modeling_llama.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -473,7 +473,7 @@ def pre_attn_forward(
use_recompute = True if os.getenv("QUANT_CONFIG", "") else False
with ht.sdp_kernel(enable_recompute=use_recompute):
attn_output = self.fused_scaled_dot_product_attention(
query_states, key_states, value_states, attention_mask, 0.0, False, None, softmax_mode
query_states, key_states, value_states, attention_mask, 0.0, False, None, "None"
)
else:
# first token
Expand Down
Loading