Skip to main content

G4eLayer

cortiq_engine::g4e

Struct G4eLayer

pub struct G4eLayer {Show 20 fields
    pub input_norm: Vec<f32>,
    pub attn_out_norm: Vec<f32>,
    pub pre_ffn_norm: Vec<f32>,
    pub post_ffw_norm: Vec<f32>,
    pub wq: QTensor,
    pub wk: Option<QTensor>,
    pub wv: Option<QTensor>,
    pub wo: QTensor,
    pub q_norm: Vec<f32>,
    pub k_norm: Option<Vec<f32>>,
    pub kv_share_src: Option<usize>,
    pub sliding: bool,
    pub hd: usize,
    pub gate: QTensor,
    pub up: QTensor,
    pub down: QTensor,
    pub ple_gate: QTensor,
    pub ple_proj: QTensor,
    pub post_ple_norm: Vec<f32>,
    pub layer_scalar: Option<f32>,
}

Fields§

§input_norm: Vec<f32>

Pre-attention norm (HF input_layernorm).

§attn_out_norm: Vec<f32>

Attention-branch output norm (HF post_attention_layernorm).

§pre_ffn_norm: Vec<f32>

Pre-FFN norm (HF pre_feedforward_layernorm).

§post_ffw_norm: Vec<f32>

FFN-branch output norm (HF post_feedforward_layernorm).

§wq: QTensor§wk: Option<QTensor>

Shared-KV layers carry no k/v projections and read the cache of kv_share_src instead of appending their own.

§wv: Option<QTensor>§wo: QTensor§q_norm: Vec<f32>§k_norm: Option<Vec<f32>>§kv_share_src: Option<usize>§sliding: bool§hd: usize

Per-layer head width: head_dim on sliding layers, global_head_dim on full layers.

§gate: QTensor§up: QTensor§down: QTensor§ple_gate: QTensor

PLE injection: gelu(gate·h) ⊙ ple → proj → norm.

§ple_proj: QTensor§post_ple_norm: Vec<f32>§layer_scalar: Option<f32>

HF layer_scalar — per-block output scale (absent = 1.0).

Auto Trait Implementations§

impl Freeze for G4eLayer

impl RefUnwindSafe for G4eLayer

impl Send for G4eLayer

impl Sync for G4eLayer

impl Unpin for G4eLayer

impl UnsafeUnpin for G4eLayer

impl UnwindSafe for G4eLayer

Blanket Implementations§

impl<T> Any for T
where T: 'static + ?Sized,

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more

impl<T> Borrow<T> for T
where T: ?Sized,

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more

impl<T> BorrowMut<T> for T
where T: ?Sized,

fn borrow_mut(&mut self) -> &mut T

Mutably borrows from an owned value. Read more

impl<T> From<T> for T

fn from(t: T) -> T

Returns the argument unchanged.

impl<T> Instrument for T

fn instrument(self, span: Span) -> Instrumented<Self> ⓘ

Instruments this type with the provided Span, returning an Instrumented wrapper. Read more

fn in_current_span(self) -> Instrumented<Self> ⓘ

Instruments this type with the current Span, returning an Instrumented wrapper. Read more

impl<T, U> Into<U> for T
where U: From<T>,

fn into(self) -> U

Calls U::from(self).

That is, this conversion is whatever the implementation of From<T> for U chooses to do.

impl<T, U> TryFrom<U> for T
where U: Into<T>,

type Error = !

The type returned in the event of a conversion error.

fn try_from(value: U) -> Result<T, !>

Performs the conversion.

impl<T, U> TryInto<U> for T
where U: TryFrom<T>,

type Error = <U as TryFrom<T>>::Error

The type returned in the event of a conversion error.

fn try_into(self) -> Result<U, <U as TryFrom<T>>::Error>

Performs the conversion.

impl<T> WithSubscriber for T

fn with_subscriber<S>(self, subscriber: S) -> WithDispatch<Self> ⓘ
where S: Into<Dispatch>,

Attaches the provided Subscriber to this type, returning a WithDispatch wrapper. Read more

fn with_current_subscriber(self) -> WithDispatch<Self> ⓘ

Attaches the current default Subscriber to this type, returning a WithDispatch wrapper. Read more