Skip to content

Fix 'unknown token ID' in HF-specific BPE implementation #162

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

Open
alan-zero wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix 'unknown token ID' in HF-specific BPE implementation #162

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
5 changes: 3 additions & 2 deletions src/hf_tokenizer.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@
#include <fstream>
#include <string>
#include <vector>
#include <limits>

// Third Party
#include <nlohmann/json.hpp>
Expand Down Expand Up @@ -475,7 +476,7 @@ Result<std::vector<uint64_t>> HFTokenizer::byte_pair_encode_(
key.c_str(),
start,
stop);
return uint64_t(0); // Return unknown token ID instead of padding
return std::numeric_limits<uint64_t>::max(); // Return unknown token ID instead of padding
}
});
}
Expand Down Expand Up @@ -521,7 +522,7 @@ std::vector<uint64_t> HFTokenizer::_byte_pair_merge(
}

uint64_t token_id = func(char_start, char_start + char_len);
if (token_id != 0) { // Assuming 0 is padding/error token
if (token_id != std::numeric_limits<uint64_t>::max()) {
word.add(token_id, char_len);
} else {
// Handle unknown character
Expand Down
Loading