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

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

Neuron 2657 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3567. Min Act: -0.0001

Data Index: 58870 (C4 (Web Text))

Max Activating Token Index: 580

Text #1

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3423. Min Act: -0.0001

Data Index: 84310 (C4 (Web Text))

Max Activating Token Index: 436

Text #2

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3236. Min Act: -0.0001

Data Index: 788118 (C4 (Web Text))

Max Activating Token Index: 433

Text #3

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3233. Min Act: -0.0001

Data Index: 827091 (C4 (Web Text))

Max Activating Token Index: 366

Text #4

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3339. Min Act: -0.0001

Data Index: 288487 (C4 (Web Text))

Max Activating Token Index: 867

Text #5

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3074. Min Act: -0.0001

Data Index: 144957 (C4 (Web Text))

Max Activating Token Index: 525

Text #6

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3037. Min Act: -0.0001

Data Index: 560937 (C4 (Web Text))

Max Activating Token Index: 536

Text #7

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3029. Min Act: -0.0001

Data Index: 155542 (C4 (Web Text))

Max Activating Token Index: 570

Text #8

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3012. Min Act: -0.0001

Data Index: 486613 (C4 (Web Text))

Max Activating Token Index: 210

Text #9

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2813. Min Act: -0.0001

Data Index: 567391 (C4 (Web Text))

Max Activating Token Index: 984

Text #10

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2690. Min Act: -0.0001

Data Index: 630915 (C4 (Web Text))

Max Activating Token Index: 723

Text #11

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2777. Min Act: -0.0001

Data Index: 278868 (C4 (Web Text))

Max Activating Token Index: 72

Text #12

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.3088. Min Act: -0.0001

Data Index: 1007485 (C4 (Web Text))

Max Activating Token Index: 830

Text #13

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2953. Min Act: -0.0001

Data Index: 467644 (C4 (Web Text))

Max Activating Token Index: 825

Text #14

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2966. Min Act: -0.0001

Data Index: 1348145 (C4 (Web Text))

Max Activating Token Index: 266

Text #15

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2852. Min Act: -0.0001

Data Index: 1287077 (C4 (Web Text))

Max Activating Token Index: 113

Text #16

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2944. Min Act: -0.0001

Data Index: 406167 (C4 (Web Text))

Max Activating Token Index: 897

Text #17

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2831. Min Act: -0.0001

Data Index: 341240 (C4 (Web Text))

Max Activating Token Index: 235

Text #18

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2808. Min Act: -0.0001

Data Index: 1221851 (C4 (Web Text))

Max Activating Token Index: 721

Text #19

Max Range: 0.3567. Min Range: -0.3567

Max Act: 0.2722. Min Act: -0.0001

Data Index: 1221314 (C4 (Web Text))

Max Activating Token Index: 705