Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 256 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1136. Min Act: -0.0002

Data Index: 708827 (C4 (Web Text))

Max Activating Token Index: 139

Text #1

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1105. Min Act: -0.0001

Data Index: 414637 (C4 (Web Text))

Max Activating Token Index: 710

Text #2

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1114. Min Act: -0.0002

Data Index: 931665 (C4 (Web Text))

Max Activating Token Index: 864

Text #3

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1078. Min Act: -0.0001

Data Index: 1248640 (C4 (Web Text))

Max Activating Token Index: 319

Text #4

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1085. Min Act: -0.0002

Data Index: 1252800 (C4 (Web Text))

Max Activating Token Index: 731

Text #5

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.1050. Min Act: -0.0002

Data Index: 138144 (C4 (Web Text))

Max Activating Token Index: 525

Text #6

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0987. Min Act: -0.0002

Data Index: 942344 (C4 (Web Text))

Max Activating Token Index: 278

Text #7

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0979. Min Act: -0.0002

Data Index: 291317 (C4 (Web Text))

Max Activating Token Index: 82

Text #8

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0959. Min Act: -0.0002

Data Index: 405278 (C4 (Web Text))

Max Activating Token Index: 1020

Text #9

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0922. Min Act: -0.0001

Data Index: 929272 (C4 (Web Text))

Max Activating Token Index: 899

Text #10

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0954. Min Act: -0.0002

Data Index: 1042952 (C4 (Web Text))

Max Activating Token Index: 742

Text #11

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0936. Min Act: -0.0002

Data Index: 1158293 (C4 (Web Text))

Max Activating Token Index: 79

Text #12

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0946. Min Act: -0.0002

Data Index: 680648 (C4 (Web Text))

Max Activating Token Index: 710

Text #13

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0922. Min Act: -0.0001

Data Index: 310761 (C4 (Web Text))

Max Activating Token Index: 106

Text #14

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0919. Min Act: -0.0002

Data Index: 675695 (C4 (Web Text))

Max Activating Token Index: 614

Text #15

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0914. Min Act: -0.0002

Data Index: 252731 (C4 (Web Text))

Max Activating Token Index: 390

Text #16

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0932. Min Act: -0.0002

Data Index: 942190 (C4 (Web Text))

Max Activating Token Index: 273

Text #17

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0913. Min Act: -0.0002

Data Index: 516275 (C4 (Web Text))

Max Activating Token Index: 795

Text #18

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0885. Min Act: -0.0002

Data Index: 1298186 (C4 (Web Text))

Max Activating Token Index: 241

Text #19

Max Range: 0.1136. Min Range: -0.1136

Max Act: 0.0884. Min Act: -0.0002

Data Index: 1253150 (C4 (Web Text))

Max Activating Token Index: 341