enum ResidualsConfigSerde {
Standard,
MultiGate {
n_stream: usize,
init_bias: f64,
per_virtual_layer: bool,
},
}Variants§
Standard
Plain additive Pre-LN residual.
MultiGate
Multi-Gate Residuals over n_stream streams.
Fields
§
init_bias: f64Initial gate bias (see MultiGateResidualConfig::init_bias).
§
per_virtual_layer: booltrue ⇒ one MGR per virtual layer; false ⇒ one per real layer
(reused across virtual passes). See MultiGate::per_virtual.
Trait Implementations§
Source§impl<'de> Deserialize<'de> for ResidualsConfigSerde
impl<'de> Deserialize<'de> for ResidualsConfigSerde
Source§fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where
__D: Deserializer<'de>,
Deserialize this value from the given Serde deserializer. Read more
Auto Trait Implementations§
impl Freeze for ResidualsConfigSerde
impl RefUnwindSafe for ResidualsConfigSerde
impl Send for ResidualsConfigSerde
impl Sync for ResidualsConfigSerde
impl Unpin for ResidualsConfigSerde
impl UnsafeUnpin for ResidualsConfigSerde
impl UnwindSafe for ResidualsConfigSerde
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more