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

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

Neuron 1615 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: 0.0115. Min Range: -0.0115

Max Act: 0.0115. Min Act: -0.0006

Data Index: 450898 (C4 (Web Text))

Max Activating Token Index: 399

Text #1

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0110. Min Act: -0.0006

Data Index: 1322062 (C4 (Web Text))

Max Activating Token Index: 315

Text #2

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0115. Min Act: -0.0006

Data Index: 1077221 (C4 (Web Text))

Max Activating Token Index: 406

Text #3

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0112. Min Act: -0.0006

Data Index: 519480 (C4 (Web Text))

Max Activating Token Index: 984

Text #4

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0113. Min Act: -0.0006

Data Index: 465930 (C4 (Web Text))

Max Activating Token Index: 901

Text #5

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0115. Min Act: -0.0006

Data Index: 79079 (C4 (Web Text))

Max Activating Token Index: 231

Text #6

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0108. Min Act: -0.0006

Data Index: 696062 (C4 (Web Text))

Max Activating Token Index: 1012

Text #7

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0113. Min Act: -0.0006

Data Index: 43937 (C4 (Web Text))

Max Activating Token Index: 419

Text #8

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0112. Min Act: -0.0006

Data Index: 696829 (C4 (Web Text))

Max Activating Token Index: 551

Text #9

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0109. Min Act: -0.0006

Data Index: 1529218 (Python Code)

Max Activating Token Index: 491

Text #10

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0111. Min Act: -0.0006

Data Index: 790443 (C4 (Web Text))

Max Activating Token Index: 186

Text #11

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0112. Min Act: -0.0006

Data Index: 1197893 (C4 (Web Text))

Max Activating Token Index: 231

Text #12

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0106. Min Act: -0.0006

Data Index: 1589213 (Python Code)

Max Activating Token Index: 102

Text #13

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0111. Min Act: -0.0006

Data Index: 77577 (C4 (Web Text))

Max Activating Token Index: 249

Text #14

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0109. Min Act: -0.0006

Data Index: 239282 (C4 (Web Text))

Max Activating Token Index: 791

Text #15

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0108. Min Act: -0.0006

Data Index: 1569157 (Python Code)

Max Activating Token Index: 30

Text #16

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0107. Min Act: -0.0006

Data Index: 1149693 (C4 (Web Text))

Max Activating Token Index: 952

Text #17

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0109. Min Act: -0.0006

Data Index: 654810 (C4 (Web Text))

Max Activating Token Index: 860

Text #18

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0106. Min Act: -0.0006

Data Index: 537242 (C4 (Web Text))

Max Activating Token Index: 361

Text #19

Max Range: 0.0115. Min Range: -0.0115

Max Act: 0.0103. Min Act: -0.0006

Data Index: 770700 (C4 (Web Text))

Max Activating Token Index: 982