mirror of
https://github.com/ikawrakow/ik_llama.cpp.git
synced 2026-01-26 17:20:01 +00:00
Fix also output
This commit is contained in:
@@ -1668,10 +1668,15 @@ static ggml_tensor * build_output(llama_context & lctx, ggml_context * ctx, ggml
|
||||
}
|
||||
cb(o.back(), "output", id);
|
||||
}
|
||||
if (o.size() == 1) cur = o.front();
|
||||
cur = ggml_concat(ctx, o[0], o[1], 0);
|
||||
for (int id = 2; id < int(o.size()); ++id) {
|
||||
cur = ggml_concat(ctx, cur, o[id], 0);
|
||||
GGML_ASSERT(!o.empty());
|
||||
if (o.size() == 1) {
|
||||
cur = o.front();
|
||||
}
|
||||
else {
|
||||
cur = ggml_concat(ctx, o[0], o[1], 0);
|
||||
for (int id = 2; id < int(o.size()); ++id) {
|
||||
cur = ggml_concat(ctx, cur, o[id], 0);
|
||||
}
|
||||
}
|
||||
} else {
|
||||
if (output_norm) {
|
||||
|
||||
Reference in New Issue
Block a user