pavlo-ruban
Results
2
comments of
pavlo-ruban
So the issue is that llama-3.2-vision models have this extra token `` with idx 128256 (0-indexed). The scores are generated for 128256 (1 token short). The actual error is index...
@miridih-jhkim11 I went with `allowed_tokens = [t for t in allowed_tokens if t < scores.shape[-1]]`, ran into graph problem when trying to compare the token like you are doing, it...