flashinfer.cake_gdn_det_train_kcp.cake_gdn_det_train_kcp_saved_bytes

flashinfer.cake_gdn_det_train_kcp.cake_gdn_det_train_kcp_saved_bytes(ctx: Any, *, H_qk: int, HV: int, save_mode: str = 'light', boundary_mode: str = 'saved', save_transfer: bool = True, with_conv: bool = True, C: int | None = None, P: int | None = None, use_qk_l2norm_in_kernel: bool = False) → int

Saved-set bytes of one layer on this rank (save_mode, save_transfer as keywords), counted per crossing fragment; use_qk_l2norm_in_kernel=True adds the FP32 r of q and k, 8 T_r H_qk bytes (boundary_mode="saved").

Parameters:
  • ctx – This rank’s context from cake_gdn_det_train_kcp_build_context().

  • H_qk (int) – Number of query and key heads.

  • HV (int) – Number of value heads.

  • **kw – Backend options for this query, including the call’s save and boundary settings.