Skip to content

server : fix incorrect num_tokens_predicted #3480

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

Merged
merged 1 commit into from
Oct 5, 2023
Merged

server : fix incorrect num_tokens_predicted #3480

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 5 additions & 3 deletions examples/server/server.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -504,9 +504,11 @@ struct llama_server_context
});
}

bool tg = true;
while (n_past < embd.size())
{
int n_eval = (int)embd.size() - n_past;
tg = n_eval == 1;
if (n_eval > params.n_batch)
{
n_eval = params.n_batch;
Expand Down Expand Up @@ -633,7 +635,9 @@ struct llama_server_context

last_n_tokens.erase(last_n_tokens.begin());
last_n_tokens.push_back(result.tok);
num_tokens_predicted++;
if (tg) {
num_tokens_predicted++;
}
}

// add it to the context
Expand Down Expand Up @@ -1124,8 +1128,6 @@ static json format_timings(llama_server_context &llama)
{
const auto timings = llama_get_timings(llama.ctx);

assert(timings.n_eval == ptrdiff_t(llama.num_tokens_predicted));

return json{
{"prompt_n", timings.n_p_eval},
{"prompt_ms", timings.t_p_eval_ms},
Expand Down