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

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

Neuron 1304 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: 0.0868. Min Range: -0.0868

Max Act: 0.0868. Min Act: -0.0002

Data Index: 893488 (C4 (Web Text))

Max Activating Token Index: 938

Text #1

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0718. Min Act: -0.0002

Data Index: 1215463 (C4 (Web Text))

Max Activating Token Index: 194

Text #2

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0715. Min Act: -0.0002

Data Index: 730046 (C4 (Web Text))

Max Activating Token Index: 9

Text #3

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0689. Min Act: -0.0002

Data Index: 1012679 (C4 (Web Text))

Max Activating Token Index: 102

Text #4

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0684. Min Act: -0.0002

Data Index: 320908 (C4 (Web Text))

Max Activating Token Index: 535

Text #5

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0622. Min Act: -0.0002

Data Index: 72716 (C4 (Web Text))

Max Activating Token Index: 690

Text #6

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0691. Min Act: -0.0001

Data Index: 552463 (C4 (Web Text))

Max Activating Token Index: 544

Text #7

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0634. Min Act: -0.0001

Data Index: 895833 (C4 (Web Text))

Max Activating Token Index: 950

Text #8

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0687. Min Act: -0.0001

Data Index: 1247665 (C4 (Web Text))

Max Activating Token Index: 1003

Text #9

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0622. Min Act: -0.0001

Data Index: 325679 (C4 (Web Text))

Max Activating Token Index: 310

Text #10

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0672. Min Act: -0.0002

Data Index: 1108265 (C4 (Web Text))

Max Activating Token Index: 164

Text #11

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0621. Min Act: -0.0002

Data Index: 1120442 (C4 (Web Text))

Max Activating Token Index: 290

Text #12

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0571. Min Act: -0.0002

Data Index: 1043515 (C4 (Web Text))

Max Activating Token Index: 716

Text #13

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0661. Min Act: -0.0002

Data Index: 84287 (C4 (Web Text))

Max Activating Token Index: 965

Text #14

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0635. Min Act: -0.0002

Data Index: 193845 (C4 (Web Text))

Max Activating Token Index: 212

Text #15

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0625. Min Act: -0.0001

Data Index: 587841 (C4 (Web Text))

Max Activating Token Index: 975

Text #16

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0623. Min Act: -0.0001

Data Index: 913287 (C4 (Web Text))

Max Activating Token Index: 993

Text #17

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0577. Min Act: -0.0001

Data Index: 1410804 (Python Code)

Max Activating Token Index: 821

Text #18

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0617. Min Act: -0.0002

Data Index: 214043 (C4 (Web Text))

Max Activating Token Index: 38

Text #19

Max Range: 0.0868. Min Range: -0.0868

Max Act: 0.0616. Min Act: -0.0002

Data Index: 67446 (C4 (Web Text))

Max Activating Token Index: 401