3Blue1Brown

How might LLMs store facts | Deep Learning Chapter 7

Examines hypotheses and partial results about how transformer multilayer perceptrons represent information.

Watch with the original provider

Choose a player when you are ready. Nothing plays automatically.

Watch at source ↗

Loading a player sends your IP address and browser information to that provider; playback may involve cookies, advertising and further data collection. Privacy and provider terms

3Blue1Brown Grant Sanderson Level not established Duration unknown

Why it is here

Examines hypotheses and partial results about how transformer multilayer perceptrons represent information.

Selection basis: editorial selection. Individual selection reason not yet checked.

Source and access

Open the original institutional or discovery page ↗

Historical watch-page observations are audit only; an authorised destination check is pending.

Access evidence
unknown · source page only
Checked
2026-11-07
Full playback observed
No
Uploaded
Unknown
Originally recorded
Unknown
Language
Unknown
Level basis
Editorial inference.
Transcript
unknown
Provenance and reuse

3Blue1Brown; original source https://www.3blue1brown.com/lessons/mlp/. Creator copyright; no blanket open licence verified.. Links, factual metadata and original navigation annotations only. No source page, transcript, image, video or restricted platform metadata is included in the reusable export.

  1. providers-source-8500ed15c00dcb74

    Official lesson text and VideoObject inspected; thumbnail recommendations and inline animation clips excluded. Retrieved 2026-11-07.

Add to a playlist