apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-12-24 13:28:13 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[BugFix] Fix flash_attn_backend

Browse Source
This commit is contained in:
lizhenyun01
2025-12-05 12:00:08 +08:00
parent 86b6430582
commit d436640735
1 changed files with 3 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
fastdeploy/model_executor/layers/attention/flash_attn_backend.py
View File

@@ -295,6 +295,8 @@ class FlashAttentionBackend(AttentionBackend):
metadata.pre_cache_batch_ids,
metadata.pre_cache_tile_ids_per_batch,
metadata.pre_cache_num_blocks_cpu,
getattr(layer, "q_norm_weight", None),
getattr(layer, "k_norm_weight", None),
getattr(layer, "cache_k_scale", None),
getattr(layer, "cache_v_scale", None),
getattr(layer, "cache_k_out_scale", None),
@@ -304,6 +306,7 @@ class FlashAttentionBackend(AttentionBackend):
metadata.kv_signal_data_list[layer.layer_id],
metadata.kv_token_num_cpu[0].item(),
self.max_seq_len,
getattr(layer, "rms_norm_eps", 1e-6),
getattr(layer, "cache_quant_type_str", "none"),
self.rope_3d,
)