Glm5NextKdaConfig

Struct Glm5NextKdaConfig 

Source
pub struct Glm5NextKdaConfig {
    pub hidden: usize,
    pub heads: usize,
    pub head_dim: usize,
    pub conv_kernel: usize,
    pub gate_lower_bound: f32,
    pub rms_norm_eps: f32,
    pub l2_eps: f32,
    pub chunk: usize,
}
Expand description

Geometry and the config values that MUST be read from the checkpoint.

gate_lower_bound, rms_norm_eps and hidden_act coincide with a library default on GLM-5.3-Flash by accident — vLLM looks up the legacy key lower_bound, misses it, and falls back to a matching -5.0; it never passes rms_norm_eps; it hardcodes "silu". Inheriting any of those defaults is a latent bug on the next checkpoint.

Fields§

§hidden: usize§heads: usize§head_dim: usize§conv_kernel: usize§gate_lower_bound: f32

linear_attn_config.gate_lower_bound.

§rms_norm_eps: f32

rms_norm_eps, consumed by o_norm.

§l2_eps: f32

FLA convention 1/sqrt(sum + eps), not max(norm, eps). Not a config key.

§chunk: usize

Prefill tiling width. A tiling parameter only — results are identical at C = 2..32, and agree with HF’s C = 64 — but bounded by SMEM_CEILING.

Implementations§

Source§

impl Glm5NextKdaConfig

Source

pub fn qkv_dim(&self) -> usize

Source

pub fn conv_dim(&self) -> usize

q | k | v concatenated: what the fused depthwise conv sees.

Source

pub fn qk_channels(&self) -> usize

Only q | k are L2-normalised. V never is.

Source

pub fn recurrent_state_elems(&self) -> usize

FP32 [heads, head_dim, head_dim], K-major. Mandatory dtype, not a policy choice.

Source

pub fn conv_state_elems(&self) -> usize

FP32 [conv_dim, conv_kernel] — Atlas’s width, one slot wider than HF’s.

Source

pub fn smem_prepare(&self) -> usize

Source

pub fn smem_scan(&self) -> usize

Source

pub fn validate(&self) -> Result<()>

Trait Implementations§

Source§

impl Clone for Glm5NextKdaConfig

Source§

fn clone(&self) -> Glm5NextKdaConfig

Returns a duplicate of the value. Read more
1.0.0 · Source§

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source. Read more
Source§

impl Debug for Glm5NextKdaConfig

Source§

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter. Read more
Source§

impl Copy for Glm5NextKdaConfig

Auto Trait Implementations§

Blanket Implementations§

Source§

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

Source§

fn type_id(&self) -> TypeId

Gets the TypeId of self. Read more
Source§

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

Source§

fn borrow(&self) -> &T

Immutably borrows from an owned value. Read more
Source§

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

Source§

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

Mutably borrows from an owned value. Read more
Source§

impl<T> CloneToUninit for T
where T: Clone,

Source§

unsafe fn clone_to_uninit(&self, dest: *mut u8)

🔬This is a nightly-only experimental API. (clone_to_uninit)
Performs copy-assignment from self to dest. Read more
Source§

impl<T> From<T> for T

Source§

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
Source§

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

Source§

fn into(self) -> U

Calls U::from(self).

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

Source§

impl<T> ToOwned for T
where T: Clone,

Source§

type Owned = T

The resulting type after obtaining ownership.
Source§

fn to_owned(&self) -> T

Creates owned data from borrowed data, usually by cloning. Read more
Source§

fn clone_into(&self, target: &mut T)

Uses borrowed data to replace owned data, usually by cloning. Read more
Source§

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

Source§

type Error = Infallible

The type returned in the event of a conversion error.
Source§

fn try_from(value: U) -> Result<T, <T as TryFrom<U>>::Error>

Performs the conversion.
Source§

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

Source§

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

The type returned in the event of a conversion error.
Source§

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

Performs the conversion.
§

impl<V, T> VZip<V> for T
where V: MultiLane<T>,

§

fn vzip(self) -> V

§

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