pub fn load_kv_scales(
store: &WeightStore,
attn_prefix: &str,
gpu: &dyn GpuBackend,
) -> Result<(f32, f32)>Expand description
2026-09-28: load_checkpoint_kv_scales, with (1.0, 1.0) for a layer the
checkpoint ships no scales for. Serve logs how many layers that is
(resolve_kv_cache_config), from the same resolver.