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

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

Neuron 309 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0127. Min Act: -0.0006

Data Index: 1229583 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0139. Min Act: -0.0006

Data Index: 1523361 (Python Code)

Max Activating Token Index: 1023

Text #2

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0148. Min Act: -0.0006

Data Index: 1601881 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0133. Min Act: -0.0006

Data Index: 215960 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0142. Min Act: -0.0006

Data Index: 1229675 (C4 (Web Text))

Max Activating Token Index: 105

Text #5

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0123. Min Act: -0.0006

Data Index: 120296 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0129. Min Act: -0.0006

Data Index: 909034 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0106. Min Act: -0.0006

Data Index: 226149 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0143. Min Act: -0.0006

Data Index: 1112329 (C4 (Web Text))

Max Activating Token Index: 1023

Text #9

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0107. Min Act: -0.0006

Data Index: 1172937 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0120. Min Act: -0.0006

Data Index: 324296 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0118. Min Act: -0.0006

Data Index: 1533829 (Python Code)

Max Activating Token Index: 1023

Text #12

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0120. Min Act: -0.0006

Data Index: 950939 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0105. Min Act: -0.0006

Data Index: 910994 (C4 (Web Text))

Max Activating Token Index: 80

Text #14

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0085. Min Act: -0.0006

Data Index: 1192859 (C4 (Web Text))

Max Activating Token Index: 685

Text #15

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0129. Min Act: -0.0006

Data Index: 1625585 (Python Code)

Max Activating Token Index: 1023

Text #16

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0117. Min Act: -0.0006

Data Index: 65966 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0128. Min Act: -0.0006

Data Index: 130213 (C4 (Web Text))

Max Activating Token Index: 190

Text #18

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0098. Min Act: -0.0006

Data Index: 390491 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.0148. Min Range: -0.0148

Max Act: 0.0101. Min Act: -0.0006

Data Index: 1522207 (Python Code)

Max Activating Token Index: 1023