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

squash! sampling : support intermixed backend/cpu samplers 

Add check that logits is not null which is can happen for embeddings.
This commit is contained in:
Daniel Bevenius 2025-11-28 08:57:48 +01:00
parent 74be332e24
commit 9ad6522be6
No known key found for this signature in database
1 changed files with 12 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

22
src/llama-context.cpp
View File

@ -1662,16 +1662,18 @@ uint32_t llama_context::output_reserve(int32_t n_outputs, const llama_batch & ba
bool batch_has_backend_sampling = false;
bool batch_needs_cpu_logits = false;
for (int32_t i = 0; i < batch.n_tokens; i++) {
if (!batch.logits[i]) {
continue;
}
for (int32_t j = 0; j < batch.n_seq_id[i]; j++) {
llama_seq_id seq_id = batch.seq_id[i][j];
if (sampling.samplers.find(seq_id) != sampling.samplers.end()) {
batch_has_backend_sampling = true;
} else {
batch_needs_cpu_logits = true;
if (batch.logits) {
for (int32_t i = 0; i < batch.n_tokens; i++) {
if (!batch.logits[i]) {
continue;
}
for (int32_t j = 0; j < batch.n_seq_id[i]; j++) {
llama_seq_id seq_id = batch.seq_id[i][j];
if (sampling.samplers.find(seq_id) != sampling.samplers.end()) {
batch_has_backend_sampling = true;
} else {
batch_needs_cpu_logits = true;
}
}
}
}