jackxlau

I was looking for a solution for this thanks!

this is neat

In my own testing I have seen peak performance happen usually within 15-20% of the intended context limit, albeit there are a few optimizations depending on the task quality.

the handshake process to unlock the A100 cores is pretty interesting it reminds me of driving CPU cores like a GPU command queue.

I came across the conclusion here since a change sometimes spans several repos, per-repo history optimizes the wrong target.