Embedding that fits the budget — pick a model that matches your corpus — step 4 of 9
Look at the four vectors in the editor. The query vector
[0.8, 0.2, 0.0, 0.0] is dominated by the first dimension —
weight 0.8 on axis 1, weak elsewhere.
Which doc shares that pattern? Predict the exact stdout: three similarity scores (4 decimal places each) and the name of the closest doc.
⌘↵ runs the editor.read, then continue.
Embedding that fits the budget — pick a model that matches your corpus — step 4 of 9
Look at the four vectors in the editor. The query vector
[0.8, 0.2, 0.0, 0.0] is dominated by the first dimension —
weight 0.8 on axis 1, weak elsewhere.
Which doc shares that pattern? Predict the exact stdout: three similarity scores (4 decimal places each) and the name of the closest doc.
⌘↵ runs the editor.read, then continue.