AI Coding · OpenAI · Hard
Problem: Repair a Transformer Implementation and Make It a Classifier A Transformer model implementation is provided, including self-attention, feed-forward layers, residual paths, LayerNorm, and related components, together with a unit-test suite. Task 1: Fix the bugs The implementation deliberately includes several defects (approximately four). Locate and correct them until every supplied test succeeds. You are free to edit any implementation detail, including attention…
Checking your access…