GeGLU

GeGLU(approximate="none")

GeGLU over a concatenated [, 2d] gate-then-up input, emitting [, d]: gelu(gate) * up.

The module form of clika_runtime.nn.functional.geglu: construct it once with the settings, then call it like a function.

Args: approximate: the GELU variant applied to the gate half.

`init`

__init__(self, approximate: '_GeluMode' = 'none') -> 'None'

Initialize self. See help(type(self)) for accurate signature.

extra_repr(self) -> 'str'

extra_repr() -> str

One line of per-class detail for :meth:__repr__; a layer prints its geometry here (in_features=64, out_features=256).

forward(self, input: 'Tensor') -> 'Tensor'

forward(input) -> Tensor

Runs clika_runtime.nn.functional.geglu on input with the stored settings; returns a new tensor.