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

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

Neuron 344 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0153. Min Act: -0.0006

Data Index: 1481893 (Python Code)

Max Activating Token Index: 170

Text #1

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0125. Min Act: -0.0006

Data Index: 1275990 (C4 (Web Text))

Max Activating Token Index: 253

Text #2

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0110. Min Act: -0.0006

Data Index: 1206876 (C4 (Web Text))

Max Activating Token Index: 508

Text #3

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0106. Min Act: -0.0006

Data Index: 740313 (C4 (Web Text))

Max Activating Token Index: 434

Text #4

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0103. Min Act: -0.0006

Data Index: 1081737 (C4 (Web Text))

Max Activating Token Index: 923

Text #5

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0101. Min Act: -0.0006

Data Index: 919088 (C4 (Web Text))

Max Activating Token Index: 623

Text #6

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0104. Min Act: -0.0006

Data Index: 1189382 (C4 (Web Text))

Max Activating Token Index: 400

Text #7

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0098. Min Act: -0.0006

Data Index: 1529838 (Python Code)

Max Activating Token Index: 369

Text #8

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0097. Min Act: -0.0005

Data Index: 84771 (C4 (Web Text))

Max Activating Token Index: 74

Text #9

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0097. Min Act: -0.0006

Data Index: 1314347 (C4 (Web Text))

Max Activating Token Index: 802

Text #10

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0096. Min Act: -0.0006

Data Index: 1386036 (Python Code)

Max Activating Token Index: 251

Text #11

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0095. Min Act: -0.0006

Data Index: 614263 (C4 (Web Text))

Max Activating Token Index: 459

Text #12

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0092. Min Act: -0.0006

Data Index: 1200251 (C4 (Web Text))

Max Activating Token Index: 88

Text #13

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0095. Min Act: -0.0005

Data Index: 326273 (C4 (Web Text))

Max Activating Token Index: 882

Text #14

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0097. Min Act: -0.0006

Data Index: 1524781 (Python Code)

Max Activating Token Index: 128

Text #15

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0097. Min Act: -0.0005

Data Index: 1471675 (Python Code)

Max Activating Token Index: 748

Text #16

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0091. Min Act: -0.0006

Data Index: 243919 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0096. Min Act: -0.0006

Data Index: 1486977 (Python Code)

Max Activating Token Index: 858

Text #18

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0097. Min Act: -0.0005

Data Index: 1481778 (Python Code)

Max Activating Token Index: 1014

Text #19

Max Range: 0.0153. Min Range: -0.0153

Max Act: 0.0092. Min Act: -0.0005

Data Index: 1528206 (Python Code)

Max Activating Token Index: 982