Embeddings

Learn in five minutes

• An embedding is a list of numbers that stands for a text.
• Cosine similarity is a comparison, not an understanding.
• Dense passage retrieval is a cited approach for finding passages this way.
• A close vector can still be the wrong fact.
• Lab 06 shows a naive neighbor search missing a keyword hit. Read that result as a lesson, not a universal law.

Keep a keyword path beside the vector path until you have measured both on your notes. The lab's ablation is one fixture, not every corpus.

This section HTML

Visual

Diagram SVGDiagram PNGDiagram HTML

Flow SVGFlow PNGFlow HTML

Two scores, one question

1. Pick a question whose words appear in only one note.
2. Score a keyword overlap and a toy vector if you have one.
3. Write which one found the note.
4. Do not generalize past that fixture.

This section HTML

Quiz

This section HTML

Ecosystem

Embedding

A numeric vector standing in for a piece of text.

Sisters: Search indexes, Embeddings, Rerankers, Citations.

A voice agent and a chat agent can share the same note index.

The capstone answers only from notes you are allowed to use.

Config

• Write the goal in one sentence.
• Name the input and the allowed output.
• Save the trace next to the result.

Troubleshooting

• The answer looks fluent and is wrong. Check whether a source passage actually contains the claim.
• The loop never stops. Set a step cap and stop when the tool result repeats.

This section HTML