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

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

Neuron 4844 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8300. Min Range: -3.8300

Max Act: 3.8300. Min Act: -0.0000

Data Index: 452533 (C4 (Web Text))

Max Activating Token Index: 863

Text #1

Max Range: 3.8300. Min Range: -3.8300

Max Act: 3.6713. Min Act: -0.0000

Data Index: 325211 (C4 (Web Text))

Max Activating Token Index: 972

Text #2

Max Range: 3.8300. Min Range: -3.8300

Max Act: 3.6321. Min Act: -0.0000

Data Index: 1288577 (C4 (Web Text))

Max Activating Token Index: 988

Text #3

Max Range: 3.8300. Min Range: -3.8300

Max Act: 3.2780. Min Act: -0.0000

Data Index: 737565 (C4 (Web Text))

Max Activating Token Index: 160

Text #4

Max Range: 3.8300. Min Range: -3.8300

Max Act: 2.7876. Min Act: -0.0000

Data Index: 35354 (C4 (Web Text))

Max Activating Token Index: 733

Text #5

Max Range: 3.8300. Min Range: -3.8300

Max Act: 2.5607. Min Act: -0.0000

Data Index: 361798 (C4 (Web Text))

Max Activating Token Index: 320

Text #6

Max Range: 3.8300. Min Range: -3.8300

Max Act: 2.1081. Min Act: -0.0000

Data Index: 276062 (C4 (Web Text))

Max Activating Token Index: 575

Text #7

Max Range: 3.8300. Min Range: -3.8300

Max Act: 2.0737. Min Act: -0.0000

Data Index: 118920 (C4 (Web Text))

Max Activating Token Index: 270

Text #8

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.8873. Min Act: -0.0000

Data Index: 810483 (C4 (Web Text))

Max Activating Token Index: 471

Text #9

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.8171. Min Act: -0.0000

Data Index: 1548457 (Python Code)

Max Activating Token Index: 790

Text #10

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7982. Min Act: -0.0000

Data Index: 387289 (C4 (Web Text))

Max Activating Token Index: 723

Text #11

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7642. Min Act: -0.0000

Data Index: 435680 (C4 (Web Text))

Max Activating Token Index: 591

Text #12

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7532. Min Act: -0.0000

Data Index: 150560 (C4 (Web Text))

Max Activating Token Index: 975

Text #13

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.8404. Min Act: -0.0000

Data Index: 135790 (C4 (Web Text))

Max Activating Token Index: 339

Text #14

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7299. Min Act: -0.0000

Data Index: 1124808 (C4 (Web Text))

Max Activating Token Index: 723

Text #15

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7149. Min Act: -0.0000

Data Index: 1178760 (C4 (Web Text))

Max Activating Token Index: 374

Text #16

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.7920. Min Act: -0.0000

Data Index: 416390 (C4 (Web Text))

Max Activating Token Index: 110

Text #17

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.6385. Min Act: -0.0000

Data Index: 1423393 (Python Code)

Max Activating Token Index: 58

Text #18

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.6071. Min Act: -0.0000

Data Index: 68709 (C4 (Web Text))

Max Activating Token Index: 783

Text #19

Max Range: 3.8300. Min Range: -3.8300

Max Act: 1.5954. Min Act: -0.0000

Data Index: 1178078 (C4 (Web Text))

Max Activating Token Index: 627