linmult.core.ffn
================

.. py:module:: linmult.core.ffn

.. autoapi-nested-parse::

   FFN residual block: two linear layers with GELU activation and residual connection.



Classes
-------

.. autoapisummary::

   linmult.core.ffn.FFNResidual


Module Contents
---------------

.. py:class:: FFNResidual(dim: int, dropout: float = 0.0)

   Bases: :py:obj:`torch.nn.Module`


   Two-layer FFN with GELU activation, dropout, and residual connection.

   Computes ``x + fc2(dropout(gelu(fc1(x))))``.

   :param dim: Input and output feature dimension.
   :type dim: int
   :param dropout: Dropout probability applied after the first linear layer.
                   Defaults to ``0.0``.
   :type dropout: float

   Initialize internal Module state, shared by both nn.Module and ScriptModule.


   .. py:method:: forward(x: torch.Tensor) -> torch.Tensor

      Apply FFN + residual.

      :param x: Input tensor of any shape with last dim ``dim``.
      :type x: torch.Tensor

      :returns: Same shape as ``x``.
      :rtype: torch.Tensor