sdpa

Function sdpa 

Source
pub fn sdpa(
    q: &[f32],
    k: &[f32],
    v: &[f32],
    t: usize,
    heads: usize,
    dq: usize,
    dv: usize,
) -> Vec<f32>
Expand description

Causal scaled-dot-product attention. q/k: [T, H, dq], v: [T, H, dv].