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

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

Neuron 2194 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.0594. Min Range: -5.0594

Max Act: 5.0594. Min Act: -0.0001

Data Index: 1214773 (C4 (Web Text))

Max Activating Token Index: 60

Text #1

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7404. Min Act: -0.0001

Data Index: 192148 (C4 (Web Text))

Max Activating Token Index: 70

Text #2

Max Range: 5.0594. Min Range: -5.0594

Max Act: 4.1210. Min Act: -0.0001

Data Index: 771911 (C4 (Web Text))

Max Activating Token Index: 679

Text #3

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.2173. Min Act: -0.0001

Data Index: 668191 (C4 (Web Text))

Max Activating Token Index: 852

Text #4

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.6198. Min Act: -0.0001

Data Index: 888387 (C4 (Web Text))

Max Activating Token Index: 370

Text #5

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.8055. Min Act: -0.0001

Data Index: 258966 (C4 (Web Text))

Max Activating Token Index: 169

Text #6

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7476. Min Act: -0.0001

Data Index: 1089775 (C4 (Web Text))

Max Activating Token Index: 61

Text #7

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.5887. Min Act: -0.0001

Data Index: 756412 (C4 (Web Text))

Max Activating Token Index: 434

Text #8

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.4111. Min Act: -0.0001

Data Index: 990222 (C4 (Web Text))

Max Activating Token Index: 100

Text #9

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.1882. Min Act: -0.0001

Data Index: 694177 (C4 (Web Text))

Max Activating Token Index: 84

Text #10

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7779. Min Act: -0.0001

Data Index: 676727 (C4 (Web Text))

Max Activating Token Index: 159

Text #11

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7113. Min Act: -0.0001

Data Index: 749206 (C4 (Web Text))

Max Activating Token Index: 115

Text #12

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.4840. Min Act: -0.0001

Data Index: 52430 (C4 (Web Text))

Max Activating Token Index: 37

Text #13

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.6288. Min Act: -0.0001

Data Index: 1276524 (C4 (Web Text))

Max Activating Token Index: 119

Text #14

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.0081. Min Act: -0.0001

Data Index: 1322610 (C4 (Web Text))

Max Activating Token Index: 922

Text #15

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7985. Min Act: -0.0001

Data Index: 312456 (C4 (Web Text))

Max Activating Token Index: 90

Text #16

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.7177. Min Act: -0.0001

Data Index: 10465 (C4 (Web Text))

Max Activating Token Index: 953

Text #17

Max Range: 5.0594. Min Range: -5.0594

Max Act: 2.8471. Min Act: -0.0001

Data Index: 660355 (C4 (Web Text))

Max Activating Token Index: 823

Text #18

Max Range: 5.0594. Min Range: -5.0594

Max Act: 3.2105. Min Act: -0.0001

Data Index: 1281648 (C4 (Web Text))

Max Activating Token Index: 91

Text #19

Max Range: 5.0594. Min Range: -5.0594

Max Act: 2.8163. Min Act: -0.0001

Data Index: 703521 (C4 (Web Text))

Max Activating Token Index: 1005