Easy

Implement next-token cross-entropy for a batch of logits/targets in numpy with padding mas

Pretraining Objectives & Scaling Laws · Problem 1 of 4

Chapter 04Pretraining Objectives & Scaling Laws

Implement next-token cross-entropy for a batch of logits/targets in numpy with padding mas

EasyProblem 1 / 4

Implement next-token cross-entropy for a batch of logits/targets in numpy with padding masking.

Implement the function/class skeleton in the editor. Any correct approach is accepted.

Hints