fix typo (llama/8267)

Co-authored-by: Judd <redacted>

diff --git a/src/ggml.c b/src/ggml.c
index f5502afbe98b36dc0c75b08b1986105057395b81..bc91ac3a726abdf08ad2cb9a76aec298c7efb61e 100644 (file)
--- a/src/ggml.c
+++ b/src/ggml.c
@@ -5312,7 +5312,7 @@ void ggml_mul_mat_set_prec(
     as  -> [cols, rows, n_expert]
     ids -> [n_experts_used, n_tokens] (i32)
     b   -> [cols, n_expert_used, n_tokens]
-    c   -> [cols, n_expert_used, n_tokens]
+    c   -> [rows, n_expert_used, n_tokens]
 
     in b, n_experts_used can be broadcasted to match the n_expert_used of ids