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

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

Neuron 2019 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: 0.0968. Min Range: -0.0968

Max Act: 0.0968. Min Act: -0.0001

Data Index: 330912 (C4 (Web Text))

Max Activating Token Index: 10

Text #1

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0864. Min Act: -0.0002

Data Index: 1183377 (C4 (Web Text))

Max Activating Token Index: 267

Text #2

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0798. Min Act: -0.0001

Data Index: 679935 (C4 (Web Text))

Max Activating Token Index: 250

Text #3

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0830. Min Act: -0.0002

Data Index: 1205923 (C4 (Web Text))

Max Activating Token Index: 219

Text #4

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0861. Min Act: -0.0002

Data Index: 474302 (C4 (Web Text))

Max Activating Token Index: 85

Text #5

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0887. Min Act: -0.0001

Data Index: 53862 (C4 (Web Text))

Max Activating Token Index: 941

Text #6

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0852. Min Act: -0.0001

Data Index: 991971 (C4 (Web Text))

Max Activating Token Index: 225

Text #7

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0842. Min Act: -0.0002

Data Index: 712748 (C4 (Web Text))

Max Activating Token Index: 416

Text #8

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0843. Min Act: -0.0002

Data Index: 306417 (C4 (Web Text))

Max Activating Token Index: 51

Text #9

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0868. Min Act: -0.0002

Data Index: 180454 (C4 (Web Text))

Max Activating Token Index: 275

Text #10

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0776. Min Act: -0.0002

Data Index: 1284337 (C4 (Web Text))

Max Activating Token Index: 810

Text #11

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0793. Min Act: -0.0002

Data Index: 1319347 (C4 (Web Text))

Max Activating Token Index: 36

Text #12

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0822. Min Act: -0.0002

Data Index: 1269264 (C4 (Web Text))

Max Activating Token Index: 72

Text #13

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0761. Min Act: -0.0002

Data Index: 1197097 (C4 (Web Text))

Max Activating Token Index: 532

Text #14

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0801. Min Act: -0.0002

Data Index: 1135695 (C4 (Web Text))

Max Activating Token Index: 733

Text #15

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0732. Min Act: -0.0002

Data Index: 1129790 (C4 (Web Text))

Max Activating Token Index: 982

Text #16

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0851. Min Act: -0.0002

Data Index: 1103048 (C4 (Web Text))

Max Activating Token Index: 560

Text #17

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0746. Min Act: -0.0001

Data Index: 941515 (C4 (Web Text))

Max Activating Token Index: 125

Text #18

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0741. Min Act: -0.0001

Data Index: 396471 (C4 (Web Text))

Max Activating Token Index: 102

Text #19

Max Range: 0.0968. Min Range: -0.0968

Max Act: 0.0741. Min Act: -0.0001

Data Index: 674158 (C4 (Web Text))

Max Activating Token Index: 695