Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1547 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0748. Min Act: -0.0002

Data Index: 123708 (C4 (Web Text))

Max Activating Token Index: 825

Text #1

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0749. Min Act: -0.0002

Data Index: 942634 (C4 (Web Text))

Max Activating Token Index: 695

Text #2

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0716. Min Act: -0.0002

Data Index: 538080 (C4 (Web Text))

Max Activating Token Index: 157

Text #3

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0649. Min Act: -0.0002

Data Index: 299094 (C4 (Web Text))

Max Activating Token Index: 379

Text #4

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0569. Min Act: -0.0002

Data Index: 1305021 (C4 (Web Text))

Max Activating Token Index: 1018

Text #5

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0530. Min Act: -0.0002

Data Index: 112398 (C4 (Web Text))

Max Activating Token Index: 189

Text #6

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0503. Min Act: -0.0001

Data Index: 60527 (C4 (Web Text))

Max Activating Token Index: 478

Text #7

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0509. Min Act: -0.0002

Data Index: 181060 (C4 (Web Text))

Max Activating Token Index: 316

Text #8

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0511. Min Act: -0.0002

Data Index: 323839 (C4 (Web Text))

Max Activating Token Index: 540

Text #9

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0451. Min Act: -0.0002

Data Index: 335933 (C4 (Web Text))

Max Activating Token Index: 173

Text #10

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0496. Min Act: -0.0001

Data Index: 981219 (C4 (Web Text))

Max Activating Token Index: 180

Text #11

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0452. Min Act: -0.0002

Data Index: 1106887 (C4 (Web Text))

Max Activating Token Index: 1019

Text #12

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0468. Min Act: -0.0001

Data Index: 280816 (C4 (Web Text))

Max Activating Token Index: 845

Text #13

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0436. Min Act: -0.0002

Data Index: 1122127 (C4 (Web Text))

Max Activating Token Index: 882

Text #14

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0444. Min Act: -0.0002

Data Index: 369561 (C4 (Web Text))

Max Activating Token Index: 152

Text #15

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0441. Min Act: -0.0001

Data Index: 1063289 (C4 (Web Text))

Max Activating Token Index: 690

Text #16

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0409. Min Act: -0.0002

Data Index: 76617 (C4 (Web Text))

Max Activating Token Index: 936

Text #17

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0448. Min Act: -0.0002

Data Index: 521940 (C4 (Web Text))

Max Activating Token Index: 763

Text #18

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0439. Min Act: -0.0002

Data Index: 1635006 (Python Code)

Max Activating Token Index: 998

Text #19

Max Range: 0.0749. Min Range: -0.0749

Max Act: 0.0433. Min Act: -0.0002

Data Index: 258219 (C4 (Web Text))

Max Activating Token Index: 524