mirror of
https://github.com/ggml-org/llama.cpp.git
synced 2026-06-27 23:50:20 -05:00
parent
d789527482
commit
8a118ee86c
@ -991,7 +991,7 @@ struct common_speculative_impl_draft_mtp : public common_speculative_impl {
|
||||
|
||||
if (chain_heads) {
|
||||
this->params.n_max = std::min(this->params.n_max, n_mtp_layers);
|
||||
|
||||
|
||||
chain_h.assign(n_seq, {});
|
||||
for (auto & c : chain_h) {
|
||||
c.reserve((size_t) (this->params.n_max + 1) * n_embd);
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user