appendK method

Tensor appendK(
  1. int kvHead,
  2. Tensor newK
)

Append [N_new, headDim] K and V for a single KV head. Returns the updated concatenated tensor (also stored in the cache).

Implementation

Tensor appendK(int kvHead, Tensor newK) {
  final prev = k[kvHead];
  final next = prev == null
      ? newK
      : TensorConcat.concat([prev, newK], axis: 0);
  k[kvHead] = next;
  return next;
}