Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4519 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.8500. Min Act: -0.0000

Data Index: 500244 (C4 (Web Text))

Max Activating Token Index: 733

Text #1

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.8201. Min Act: -0.0000

Data Index: 1064038 (C4 (Web Text))

Max Activating Token Index: 818

Text #2

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.7696. Min Act: -0.0000

Data Index: 81629 (C4 (Web Text))

Max Activating Token Index: 913

Text #3

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.6676. Min Act: -0.0000

Data Index: 1134411 (C4 (Web Text))

Max Activating Token Index: 21

Text #4

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.6309. Min Act: -0.0000

Data Index: 16650 (C4 (Web Text))

Max Activating Token Index: 82

Text #5

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.6208. Min Act: -0.0000

Data Index: 956535 (C4 (Web Text))

Max Activating Token Index: 27

Text #6

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.5697. Min Act: -0.0000

Data Index: 736677 (C4 (Web Text))

Max Activating Token Index: 823

Text #7

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4898. Min Act: -0.0000

Data Index: 677942 (C4 (Web Text))

Max Activating Token Index: 707

Text #8

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.5555. Min Act: -0.0000

Data Index: 601226 (C4 (Web Text))

Max Activating Token Index: 805

Text #9

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4409. Min Act: -0.0000

Data Index: 346080 (C4 (Web Text))

Max Activating Token Index: 134

Text #10

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4962. Min Act: -0.0000

Data Index: 460081 (C4 (Web Text))

Max Activating Token Index: 19

Text #11

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4588. Min Act: -0.0000

Data Index: 1394913 (Python Code)

Max Activating Token Index: 33

Text #12

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4542. Min Act: -0.0000

Data Index: 138192 (C4 (Web Text))

Max Activating Token Index: 507

Text #13

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4176. Min Act: -0.0000

Data Index: 1606941 (Python Code)

Max Activating Token Index: 154

Text #14

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4503. Min Act: -0.0000

Data Index: 378014 (C4 (Web Text))

Max Activating Token Index: 1021

Text #15

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4066. Min Act: -0.0000

Data Index: 1284939 (C4 (Web Text))

Max Activating Token Index: 435

Text #16

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4511. Min Act: -0.0000

Data Index: 695574 (C4 (Web Text))

Max Activating Token Index: 772

Text #17

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4168. Min Act: -0.0000

Data Index: 1411122 (Python Code)

Max Activating Token Index: 220

Text #18

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4526. Min Act: -0.0000

Data Index: 1029332 (C4 (Web Text))

Max Activating Token Index: 774

Text #19

Max Range: 0.8500. Min Range: -0.8500

Max Act: 0.4281. Min Act: -0.0000

Data Index: 31231 (C4 (Web Text))

Max Activating Token Index: 676