upstream-archive/llama.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2026-10-02 19:07:25 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix v granularity

Browse Source
This commit is contained in:
Aman Gupta
2026-09-24 02:02:53 +08:00
parent c23433c001
commit 96335fc14f
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+1 -1
View File
@@ -768,7 +768,7 @@ struct ggml_backend_meta_split_state llama_meta_device_get_split_state(const str
}
if (std::regex_match(tensor_name, pattern_attn_out_weight)) {
GGML_ASSERT(segments.size() == 1);
return {granularity_q};
return {granularity_head * hparams.n_embd_head_v(il)};
}
if (std::regex_match(tensor_name, pattern_attn_gate_weight)) {
GGML_ASSERT(segments.size() == 1);