Transformer: Difference between revisions
Jump to navigation
Jump to search
skeleton |
No edit summary |
||
| Line 3: | Line 3: | ||
== Intuition == | == Intuition == | ||
Rongzhou's [[User:Rongzhou/Transformer-Intuition|Transformer is a super charged next word prediction algorithm]] provides a short intuition for Transformers. | Rongzhou's ''[[User:Rongzhou/Transformer-Intuition|Transformer is a super charged next word prediction algorithm]]'' provides a short intuition for Transformers. | ||
Interestingly, as [[User:Rongzhou/BehindScenes/Transformer-Intuition|the author explains himself]], the story | Interestingly, as [[User:Rongzhou/BehindScenes/Transformer-Intuition|the author explains himself]], the story doubles as a live demonstration of Transformers, as parts of the story was co-authored by a Transformer model. | ||
== Mathematical expression == | == Mathematical expression == | ||
Revision as of 17:25, 22 August 2026
A Transformer is, at the time of writing(2026-08), the most popular architecture for building Large Language Models.
Intuition
Rongzhou's Transformer is a super charged next word prediction algorithm provides a short intuition for Transformers.
Interestingly, as the author explains himself, the story doubles as a live demonstration of Transformers, as parts of the story was co-authored by a Transformer model.