graduate_code/starters/attention.py
# Exercise scaffold. Reference implementation is in a separate directory.
"""G14: single-head scaled dot-product attention, not a trained transformer."""
import math
def softmax(scores, allowed=None):
raise NotImplementedError('Implement this contract from the book.')
def attention(Q, K, V, causal=False):
raise NotImplementedError('Implement this contract from the book.')