From 07e9ec2b66056d7e73854ae78bbe3783fdd66604 Mon Sep 17 00:00:00 2001 From: Mohit Sharma Date: Tue, 10 Dec 2024 10:54:52 +0000 Subject: [PATCH] update partition size --- server/text_generation_server/layers/attention/rocm.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/server/text_generation_server/layers/attention/rocm.py b/server/text_generation_server/layers/attention/rocm.py index d65054a1..a401b589 100644 --- a/server/text_generation_server/layers/attention/rocm.py +++ b/server/text_generation_server/layers/attention/rocm.py @@ -179,7 +179,7 @@ def paged_attention( 1.0, 1.0, None, - 512, + _PARTITION_SIZE, ) return out