public step(output as NDArray<Float>, qkv as NDArray<Float>, gate as NDArray<Float>, alpha as NDArray<Float>, beta as NDArray<Float>, conv as DeviceQuantizedMatrix, bias as NDArray<Float>, decay as NDArray<Float>, norm as NDArray<Float>) as Void
Updates convolution and recurrence, then applies RMS normalization/SiLU gate.
| Parameter | Description |
|---|---|
output | Float [1, valueHeads*valueDim] destination. |
qkv | Mixed projection before convolution. |
gate | SiLU output gate projection. |
alpha | Time-step projection. |
beta | Update projection. |
conv | Packed causal depthwise weights. |
bias | Time-step bias. |
decay | Log decay weights. |
norm | Value-width normalization weights, without zero centering. |
state.step(y, qkv, z, a, b, conv, bias, decay, norm)