Your next step

Representing relationships with numbers

Learned rows, similarity, shares and weighted mixing.

What you’ll learn

A vector is a list of numbers
Read a small labeled numeric representation.
About 4 minutes · Open activity
An embedding is a learned representation
Explain why a token ID is mapped to a numeric vector.
About 4 minutes · Open activity
Nearness can make comparison useful
Interpret closeness in an illustrative embedding map.
About 4 minutes · Open activity
The axes are not usually named ideas
Recognize limits of a two-dimensional embedding illustration.
About 4 minutes · Open activity
Context can change the intended meaning
Identify which surrounding words disambiguate a token.
About 4 minutes · Open activity
A score is a degree of match
Compare supplied numerical match scores without treating them as truth probabilities.
About 4 minutes · Open activity
Turn scores into shares
Interpret normalized nonnegative weights that add to one.
About 4 minutes · Open activity
Representations can improve through training
Identify an embedding as part of a learned system.
About 4 minutes · Open activity
Mix information in unequal amounts
Predict which value contributes most to a weighted combination.
About 4 minutes · Open activity
Read a relationship map
Interpret a toy embedding map, distinguish a lookup ID from its vector, and predict the dominant contribution in a weighted mix.
About 6 minutes · Open activity