Skip to content

sdpa: support attn_mask.requires_grad, support expanded number of heads in attn_mask #5440

sdpa: support attn_mask.requires_grad, support expanded number of heads in attn_mask

sdpa: support attn_mask.requires_grad, support expanded number of heads in attn_mask #5440

docs-make (doctest)

succeeded Dec 17, 2024 in 2m 40s