ClassroomThe classroom
Back to the lesson libraryMECHANISM · 13 MIN
M10.8 CONNECT THE MECHANISM

Find the first broken assumption in a training run

Training bugs rarely crash. They just make a model quietly worse. Learn the checks that find them fast, from a two-minute sanity test to a symptom-by-symptom field guide.

LESSON OVERVIEW13 min lesson

Lesson overview

Training bugs rarely crash. They just make a model quietly worse. Learn the checks that find them fast, from a two-minute sanity test to a symptom-by-symptom field guide.

What you’ll explore

  • Training diagnosis checks data, shapes, loss, gradients, updates, and evaluation in order; divergence, plateaus, and noisy metrics can have different causes.
GO TO THE SOURCE

Original explanations, connected to the research.

Dive into Deep Learning — authors’ open textbookA Recipe for Training Neural Networks (Andrej Karpathy, 2019)
Suggest a correction

A precise note can make an explanation better.

Choose the scene and describe what needs attention. Download a feedback file to share through a channel you already use. This page does not send feedback or connect you with a reviewer.

The file includes this note, the scene title, and lesson metadata. Your saved progress and quiz responses are excluded. Download before leaving or reloading to keep your note.