Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 1538 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0813. Min Act: -0.0001

Data Index: 382016 (C4 (Web Text))

Max Activating Token Index: 706

Text #1

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0646. Min Act: -0.0001

Data Index: 986249 (C4 (Web Text))

Max Activating Token Index: 33

Text #2

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0661. Min Act: -0.0001

Data Index: 1357085 (C4 (Web Text))

Max Activating Token Index: 294

Text #3

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0580. Min Act: -0.0001

Data Index: 946884 (C4 (Web Text))

Max Activating Token Index: 46

Text #4

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0596. Min Act: -0.0001

Data Index: 218484 (C4 (Web Text))

Max Activating Token Index: 52

Text #5

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0572. Min Act: -0.0001

Data Index: 1073687 (C4 (Web Text))

Max Activating Token Index: 160

Text #6

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0548. Min Act: -0.0002

Data Index: 43559 (C4 (Web Text))

Max Activating Token Index: 528

Text #7

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0560. Min Act: -0.0001

Data Index: 1042478 (C4 (Web Text))

Max Activating Token Index: 491

Text #8

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0541. Min Act: -0.0002

Data Index: 910887 (C4 (Web Text))

Max Activating Token Index: 61

Text #9

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0554. Min Act: -0.0001

Data Index: 574000 (C4 (Web Text))

Max Activating Token Index: 262

Text #10

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0545. Min Act: -0.0001

Data Index: 196400 (C4 (Web Text))

Max Activating Token Index: 879

Text #11

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0578. Min Act: -0.0002

Data Index: 226702 (C4 (Web Text))

Max Activating Token Index: 614

Text #12

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0520. Min Act: -0.0002

Data Index: 379997 (C4 (Web Text))

Max Activating Token Index: 227

Text #13

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0520. Min Act: -0.0001

Data Index: 735439 (C4 (Web Text))

Max Activating Token Index: 883

Text #14

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0548. Min Act: -0.0002

Data Index: 355840 (C4 (Web Text))

Max Activating Token Index: 134

Text #15

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0592. Min Act: -0.0001

Data Index: 557526 (C4 (Web Text))

Max Activating Token Index: 347

Text #16

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0554. Min Act: -0.0001

Data Index: 965349 (C4 (Web Text))

Max Activating Token Index: 789

Text #17

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0569. Min Act: -0.0001

Data Index: 1228826 (C4 (Web Text))

Max Activating Token Index: 805

Text #18

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0535. Min Act: -0.0002

Data Index: 778041 (C4 (Web Text))

Max Activating Token Index: 335

Text #19

Max Range: 0.0813. Min Range: -0.0813

Max Act: 0.0522. Min Act: -0.0001

Data Index: 259808 (C4 (Web Text))

Max Activating Token Index: 564