CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-10-30 08:42:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

server : send token probs for "stream == false" (#4714)

Browse Source
This commit is contained in:
Georgi Gerganov
2024-01-04 19:56:33 +02:00
committed by GitHub
parent a91928014f
commit 012cf349ae
1 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
examples/server/server.cpp
View File

@@ -1265,7 +1265,7 @@ struct llama_server_context
{ {
std::vector<completion_token_output> probs_output = {}; std::vector<completion_token_output> probs_output = {};
const std::vector<llama_token> to_send_toks = llama_tokenize(ctx, tkn.text_to_send, false); const std::vector<llama_token> to_send_toks = llama_tokenize(ctx, tkn.text_to_send, false);
size_t probs_pos = std::min(slot.sent_token_probs_index, slot.generated_token_probs.size()); size_t probs_pos = std::min(slot.sent_token_probs_index, slot.generated_token_probs.size());
size_t probs_stop_pos = std::min(slot.sent_token_probs_index + to_send_toks.size(), slot.generated_token_probs.size()); size_t probs_stop_pos = std::min(slot.sent_token_probs_index + to_send_toks.size(), slot.generated_token_probs.size());
if (probs_pos < probs_stop_pos) if (probs_pos < probs_stop_pos)
{ {
@@ -1325,7 +1325,7 @@ struct llama_server_context
{ {
probs = std::vector<completion_token_output>( probs = std::vector<completion_token_output>(
slot.generated_token_probs.begin(), slot.generated_token_probs.begin(),
slot.generated_token_probs.begin() + slot.sent_token_probs_index); slot.generated_token_probs.end());
} }
res.result_json["completion_probabilities"] = probs_vector_to_json(ctx, probs); res.result_json["completion_probabilities"] = probs_vector_to_json(ctx, probs);
} }