Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1340 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.9884. Min Act: -0.0002

Data Index: 770231 (C4 (Web Text))

Max Activating Token Index: 6

Text #1

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.9366. Min Act: -0.0002

Data Index: 495476 (C4 (Web Text))

Max Activating Token Index: 2

Text #2

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.9366. Min Act: -0.0002

Data Index: 1215509 (C4 (Web Text))

Max Activating Token Index: 2

Text #3

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.8378. Min Act: -0.0002

Data Index: 477895 (C4 (Web Text))

Max Activating Token Index: 3

Text #4

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.8706. Min Act: -0.0002

Data Index: 893806 (C4 (Web Text))

Max Activating Token Index: 28

Text #5

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.8000. Min Act: -0.0002

Data Index: 1165628 (C4 (Web Text))

Max Activating Token Index: 6

Text #6

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7257. Min Act: -0.0002

Data Index: 1262357 (C4 (Web Text))

Max Activating Token Index: 8

Text #7

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.6936. Min Act: -0.0002

Data Index: 715232 (C4 (Web Text))

Max Activating Token Index: 33

Text #8

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7299. Min Act: -0.0002

Data Index: 177751 (C4 (Web Text))

Max Activating Token Index: 245

Text #9

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7168. Min Act: -0.0002

Data Index: 354819 (C4 (Web Text))

Max Activating Token Index: 3

Text #10

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7331. Min Act: -0.0002

Data Index: 1245488 (C4 (Web Text))

Max Activating Token Index: 5

Text #11

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.6494. Min Act: -0.0002

Data Index: 611828 (C4 (Web Text))

Max Activating Token Index: 7

Text #12

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.6536. Min Act: -0.0002

Data Index: 422643 (C4 (Web Text))

Max Activating Token Index: 298

Text #13

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7244. Min Act: -0.0002

Data Index: 1561290 (Python Code)

Max Activating Token Index: 3

Text #14

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.7024. Min Act: -0.0002

Data Index: 88125 (C4 (Web Text))

Max Activating Token Index: 4

Text #15

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.6533. Min Act: -0.0002

Data Index: 667386 (C4 (Web Text))

Max Activating Token Index: 4

Text #16

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.5930. Min Act: -0.0002

Data Index: 1386592 (Python Code)

Max Activating Token Index: 38

Text #17

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.5933. Min Act: -0.0002

Data Index: 1044096 (C4 (Web Text))

Max Activating Token Index: 21

Text #18

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.5348. Min Act: -0.0003

Data Index: 139755 (C4 (Web Text))

Max Activating Token Index: 889

Text #19

Max Range: 1.9884. Min Range: -1.9884

Max Act: 1.5776. Min Act: -0.0002

Data Index: 861339 (C4 (Web Text))

Max Activating Token Index: 305