tqcq/llama.cpp
0
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-06-27 20:05:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

server : use common_token_to_piece instead of common_detokenize (#11740)

Browse Source 
* server : use common_token_to_piece instead of common_detokenize

This commit replaces the call to common_detokenize with
common_token_to_piece in the populate_token_probs.

The motivation for this change is to avoid an issue where
common_detokenize would remove the word boundary character for tokens,
which caused a regression in the server generated token probabilities.

Resolves: https://github.com/ggerganov/llama.cpp/issues/11728

* squash! server : use common_token_to_piece instead of common_detokenize

Use common_token_to_piece for post_sampling_probs as well.
This commit is contained in:
Daniel Bevenius
2025-02-11 14:06:45 +01:00
committed by GitHub
parent b9ab0a4d0b
commit a18f481f99
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

@ -2279,7 +2279,7 @@ struct server_context {
for (size_t i = 0; i < std::min(max_probs, n_probs); i++) { for (size_t i = 0; i < std::min(max_probs, n_probs); i++) {
result.probs.push_back({ result.probs.push_back({
cur_p->data[i].id, cur_p->data[i].id,
common_detokenize(ctx, {cur_p->data[i].id}, special), common_token_to_piece(ctx, cur_p->data[i].id, special),
cur_p->data[i].p cur_p->data[i].p
}); });
} }
@ -2301,7 +2301,7 @@ struct server_context {
for (size_t i = 0; i < std::min(n_vocab, n_probs); i++) { for (size_t i = 0; i < std::min(n_vocab, n_probs); i++) {
result.probs.push_back({ result.probs.push_back({
cur[i].id, cur[i].id,
common_detokenize(ctx, {cur[i].id}, special), common_token_to_piece(ctx, cur[i].id, special),
cur[i].p cur[i].p
}); });
} }