Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Support special tokens as reverse/anti prompt. #5847

Merged
merged 3 commits into from
Mar 4, 2024
Merged

Support special tokens as reverse/anti prompt. #5847

Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
ef65124
Support special tokens as reverse/anti prompt.
dranger003 Mar 3, 2024
60325ec
Tokenize antiprompts only once.
dranger003 Mar 3, 2024
98dae32
main : minor
ggerganov Mar 4, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
3 changes: 2 additions & 1 deletion examples/main/main.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -760,7 +760,8 @@ int main(int argc, char ** argv) {
? last_output.length() - static_cast<size_t>(antiprompt.length() + extra_padding)
: 0;

if (last_output.find(antiprompt, search_start_pos) != std::string::npos) {
auto tmp = ::llama_tokenize(ctx, antiprompt, false, true);
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe pre-tokenize once at the start of the program, to avoid tokenization overhead in each iteration

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think it should be better now.

if (last_output.find(antiprompt, search_start_pos) != std::string::npos || (tmp.size() == 1 && llama_sampling_last(ctx_sampling) == tmp[0])) {
if (params.interactive) {
is_interacting = true;
}
Expand Down
Loading