Commit 89f9080e authored by David Majnemer's avatar David Majnemer Committed by TensorFlower Gardener
Browse files

[XLA] Support strength reducing bfloat16 dot products 

There is no reason to limit ourselves to float32 dot product operations, we
simply convert to and from float32 around the reduction to simulate the
precision change.

PiperOrigin-RevId: 213014410
parent 9445d19a
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment