transformers
Fix CUDA index out of bounds for q_idx in VLM token type masking for Gemma3, PaliGemma, and example modular
#41757
Merged
Merged
Fix CUDA index out of bounds for q_idx in VLM token type masking for Gemma3, PaliGemma, and example modular #41757
zucchini-nlp
merged
5 commits into
huggingface:main
from
albertvillanova:fix-cuda-error-index-out-of-bounds
Fix CUDA index out of bounds for q_idx in Gemma3 token type masking
f8e5c2a4
Fix CUDA index out of bounds for q_idx in modular modeling_new_task_m…
0e3f93fe
Revert "Fix CUDA index out of bounds for q_idx in Gemma3 token type m…
4da945f5
Fix CUDA index out of bounds for q_idx in PaliGemma token type masking
df7298a7
Fix CUDA index out of bounds for q_idx in Gemma3 token type masking
ff6af300
albertvillanova
changed the title 
zucchini-nlp
commented
on 2025-10-21
zucchini-nlp
approved these changes
on 2025-10-22
Login to write a write a comment.
Login via GitHub