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

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

Neuron 193 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0139. Min Act: -0.0006

Data Index: 26242 (C4 (Web Text))

Max Activating Token Index: 955

Text #1

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0124. Min Act: -0.0006

Data Index: 16880 (C4 (Web Text))

Max Activating Token Index: 631

Text #2

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0124. Min Act: -0.0006

Data Index: 586391 (C4 (Web Text))

Max Activating Token Index: 955

Text #3

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0126. Min Act: -0.0006

Data Index: 1303692 (C4 (Web Text))

Max Activating Token Index: 383

Text #4

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0114. Min Act: -0.0006

Data Index: 105192 (C4 (Web Text))

Max Activating Token Index: 627

Text #5

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0113. Min Act: -0.0006

Data Index: 732181 (C4 (Web Text))

Max Activating Token Index: 71

Text #6

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0101. Min Act: -0.0006

Data Index: 896032 (C4 (Web Text))

Max Activating Token Index: 806

Text #7

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0098. Min Act: -0.0006

Data Index: 1339714 (C4 (Web Text))

Max Activating Token Index: 787

Text #8

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0094. Min Act: -0.0006

Data Index: 133470 (C4 (Web Text))

Max Activating Token Index: 144

Text #9

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0091. Min Act: -0.0006

Data Index: 45211 (C4 (Web Text))

Max Activating Token Index: 830

Text #10

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0093. Min Act: -0.0006

Data Index: 244177 (C4 (Web Text))

Max Activating Token Index: 398

Text #11

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0093. Min Act: -0.0006

Data Index: 312479 (C4 (Web Text))

Max Activating Token Index: 625

Text #12

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0090. Min Act: -0.0006

Data Index: 758194 (C4 (Web Text))

Max Activating Token Index: 252

Text #13

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0087. Min Act: -0.0006

Data Index: 1443849 (Python Code)

Max Activating Token Index: 781

Text #14

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0084. Min Act: -0.0006

Data Index: 119309 (C4 (Web Text))

Max Activating Token Index: 851

Text #15

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0085. Min Act: -0.0006

Data Index: 1173228 (C4 (Web Text))

Max Activating Token Index: 591

Text #16

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0084. Min Act: -0.0006

Data Index: 948592 (C4 (Web Text))

Max Activating Token Index: 546

Text #17

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0084. Min Act: -0.0006

Data Index: 797580 (C4 (Web Text))

Max Activating Token Index: 491

Text #18

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0083. Min Act: -0.0006

Data Index: 1250166 (C4 (Web Text))

Max Activating Token Index: 605

Text #19

Max Range: 0.0139. Min Range: -0.0139

Max Act: 0.0082. Min Act: -0.0006

Data Index: 910338 (C4 (Web Text))

Max Activating Token Index: 313