Add Flax FAQ - how to search, how to take the derivative w.r.t. a hidden layer, remat_scan vs scan(remat), recommended training libraries/metrics

[8bitmp3]

This PR adds a new Flax FAQ.

• Item 1: How to search
• Item 2: How to take a derivative w.r.t. a hidden layer with flax.linen.Module.perturb
• Item 3: Flax Linen remat_scan vs. scan(remat(...))
• Item 4: Recommended training loop libraries (CLU, etc)

[codecov-commenter]

Codecov Report

Merging #3267 (7b1d8a6) into main (6635436) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main    #3267   +/-   ##
=======================================
  Coverage   83.60%   83.60%           
=======================================
  Files          56       56           
  Lines        6747     6747           
=======================================
  Hits         5641     5641           
  Misses       1106     1106           

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB