Add tip for non-CUDA users (#2434)

Author: micahscopes

docs/old_tutorials/2024-04-10-blitz.md

@@ -365,6 +365,8 @@ Seeing our training routine unfold gives us an idea of how the network learnt th
 
 The `gpu` functions you see sprinkled through this bit of the code tell Flux to move these entities to an available GPU, and subsequently train on it. No extra faffing about required! The same bit of code would work on any hardware with some small annotations like you saw here.
 
+If you're not using `CUDA`, be sure to [configure Flux.jl for your GPU backend](https://fluxml.ai/Flux.jl/stable/gpu/#Selecting-GPU-backend).
+
 ### Testing the Network
 
 We have trained the network for 100 passes over the training dataset. But we need to check if the network has learnt anything at all.