Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 258 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.1026. Min Act: -0.0006

Data Index: 1029649 (C4 (Web Text))

Max Activating Token Index: 245

Text #1

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0995. Min Act: -0.0006

Data Index: 648007 (C4 (Web Text))

Max Activating Token Index: 386

Text #2

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0950. Min Act: -0.0005

Data Index: 1128103 (C4 (Web Text))

Max Activating Token Index: 378

Text #3

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0945. Min Act: -0.0006

Data Index: 1258995 (C4 (Web Text))

Max Activating Token Index: 253

Text #4

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0921. Min Act: -0.0006

Data Index: 167047 (C4 (Web Text))

Max Activating Token Index: 105

Text #5

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0864. Min Act: -0.0006

Data Index: 839289 (C4 (Web Text))

Max Activating Token Index: 419

Text #6

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0862. Min Act: -0.0006

Data Index: 1060144 (C4 (Web Text))

Max Activating Token Index: 373

Text #7

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0791. Min Act: -0.0006

Data Index: 1600409 (Python Code)

Max Activating Token Index: 887

Text #8

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0840. Min Act: -0.0005

Data Index: 907497 (C4 (Web Text))

Max Activating Token Index: 859

Text #9

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0831. Min Act: -0.0006

Data Index: 987256 (C4 (Web Text))

Max Activating Token Index: 607

Text #10

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0791. Min Act: -0.0006

Data Index: 1027144 (C4 (Web Text))

Max Activating Token Index: 439

Text #11

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0831. Min Act: -0.0005

Data Index: 782979 (C4 (Web Text))

Max Activating Token Index: 53

Text #12

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0813. Min Act: -0.0005

Data Index: 118334 (C4 (Web Text))

Max Activating Token Index: 413

Text #13

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0829. Min Act: -0.0006

Data Index: 1043506 (C4 (Web Text))

Max Activating Token Index: 1019

Text #14

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0815. Min Act: -0.0006

Data Index: 487597 (C4 (Web Text))

Max Activating Token Index: 64

Text #15

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0774. Min Act: -0.0005

Data Index: 550981 (C4 (Web Text))

Max Activating Token Index: 665

Text #16

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0768. Min Act: -0.0005

Data Index: 1258813 (C4 (Web Text))

Max Activating Token Index: 329

Text #17

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0796. Min Act: -0.0005

Data Index: 1026222 (C4 (Web Text))

Max Activating Token Index: 541

Text #18

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0786. Min Act: -0.0005

Data Index: 1081372 (C4 (Web Text))

Max Activating Token Index: 893

Text #19

Max Range: 0.1026. Min Range: -0.1026

Max Act: 0.0758. Min Act: -0.0006

Data Index: 155215 (C4 (Web Text))

Max Activating Token Index: 15