depth must be less than the rank of `tensor`: `indices.shape[-1] <= tensor.ndim`

if `indices.shape[-1] = tensor.rank` this Op indexes and updates scalar elements.
if `indices.shape[-1] < tensor.rank` it indexes and updates slices of the input
`tensor`.

Each `update` has a rank of `tensor.rank - indices.shape[-1]`.
The overall shape of `updates` is:

```

          "start": {
            "default": 0,

          "field": {

*   Enable JIT-compiled i64-indexed kernels on GPU for large tensors with more than 2**32 elements.
    *   Unary GPU kernels: Abs, Atanh, Acos, Acosh, Asin, Asinh, Atan, Cos, Cosh, Sin, Sinh, Tan, Tanh.