For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).
Python function
make_interleaved_gated_activation_fn
make_interleaved_gated_activation_fn()
max.nn.moe.make_interleaved_gated_activation_fn(activation_fn)
Builds a gated activation for interleaved [g0, u0, g1, u1, ...]
projections.
-
Parameters:
-
activation_fn (Callable[[TensorValue], TensorValue])
-
Return type: