happyz
/
llama.cpp
mirror of https://github.com/ggerganov/llama.cpp.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

ggml : remove KQ mask padding

This commit is contained in:
Georgi Gerganov 2025-09-28 18:10:17 +03:00
parent 46c338fc6c
commit 837b1b4563
No known key found for this signature in database
GPG Key ID: 449E073F9DC10735
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ggml/include/ggml.h
View File

@ -2152,7 +2152,7 @@ extern "C" {
struct ggml_tensor * a, struct ggml_tensor * a,
int k); int k);
#define GGML_KQ_MASK_PAD 64 #define GGML_KQ_MASK_PAD 1
// q: [n_embd_k, n_batch, n_head, ne3 ] // q: [n_embd_k, n_batch, n_head, ne3 ]
// k: [n_embd_k, n_kv, n_head_kv, ne3 ] // k: [n_embd_k, n_kv, n_head_kv, ne3 ]