vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

gguf-py : fail fast on nonsensical special token IDs (#4489)

Browse source
This commit is contained in:
Jared Van Bortel 2023-12-17 10:45:46 -05:00 • committed by GitHub
parent 919c40660f
commit f7f468a97d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

4
gguf-py/gguf/vocab.py
View file

@ -109,8 +109,10 @@ class SpecialVocab:
return True return True
def _set_special_token(self, typ: str, tid: Any) -> None: def _set_special_token(self, typ: str, tid: Any) -> None:
if not isinstance(tid, int) or tid < 0: if not isinstance(tid, int):
return return
if tid < 0:
raise ValueError(f'invalid value for special token type {typ}: {tid}')
if self.n_vocab is None or tid < self.n_vocab: if self.n_vocab is None or tid < self.n_vocab:
if typ in self.special_token_ids: if typ in self.special_token_ids:
return return