Exercises
LM-C05-C-001
Split restartable as re start able. How many tokens are produced?
Compute it first, then check your number.
LM-C05-C-002
In a b a b a b, how many times does the adjacent pair a b occur?
Compute it first, then check your number.
LM-C05-C-003
After merging a b -> ab, how many tokens are in a b a b?
Compute it first, then check your number.
LM-C05-C-004
A sequence of length 9 is truncated to length 6. How many tokens are removed?
Compute it first, then check your number.
LM-C05-C-005
A vocabulary has 32,000 tokens. How many rows does its token-embedding matrix have?
Compute it first, then check your number.