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

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

Neuron 150 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: 0.0141. Min Range: -0.0141

Max Act: 0.0137. Min Act: -0.0006

Data Index: 894509 (C4 (Web Text))

Max Activating Token Index: 316

Text #1

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0141. Min Act: -0.0006

Data Index: 1069296 (C4 (Web Text))

Max Activating Token Index: 1015

Text #2

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0128. Min Act: -0.0006

Data Index: 1629518 (Python Code)

Max Activating Token Index: 1023

Text #3

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0131. Min Act: -0.0006

Data Index: 255538 (C4 (Web Text))

Max Activating Token Index: 700

Text #4

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0127. Min Act: -0.0006

Data Index: 1556230 (Python Code)

Max Activating Token Index: 809

Text #5

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0115. Min Act: -0.0006

Data Index: 268203 (C4 (Web Text))

Max Activating Token Index: 379

Text #6

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0118. Min Act: -0.0006

Data Index: 1355648 (C4 (Web Text))

Max Activating Token Index: 704

Text #7

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0113. Min Act: -0.0006

Data Index: 791628 (C4 (Web Text))

Max Activating Token Index: 914

Text #8

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0115. Min Act: -0.0006

Data Index: 867081 (C4 (Web Text))

Max Activating Token Index: 152

Text #9

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0113. Min Act: -0.0005

Data Index: 1372483 (Python Code)

Max Activating Token Index: 694

Text #10

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0107. Min Act: -0.0006

Data Index: 740816 (C4 (Web Text))

Max Activating Token Index: 186

Text #11

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0107. Min Act: -0.0005

Data Index: 676635 (C4 (Web Text))

Max Activating Token Index: 446

Text #12

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0108. Min Act: -0.0006

Data Index: 315841 (C4 (Web Text))

Max Activating Token Index: 989

Text #13

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0095. Min Act: -0.0006

Data Index: 439987 (C4 (Web Text))

Max Activating Token Index: 457

Text #14

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0106. Min Act: -0.0006

Data Index: 165467 (C4 (Web Text))

Max Activating Token Index: 184

Text #15

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0101. Min Act: -0.0006

Data Index: 583658 (C4 (Web Text))

Max Activating Token Index: 910

Text #16

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0110. Min Act: -0.0005

Data Index: 1399163 (Python Code)

Max Activating Token Index: 1023

Text #17

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0103. Min Act: -0.0006

Data Index: 1354689 (C4 (Web Text))

Max Activating Token Index: 707

Text #18

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0110. Min Act: -0.0006

Data Index: 1318578 (C4 (Web Text))

Max Activating Token Index: 916

Text #19

Max Range: 0.0141. Min Range: -0.0141

Max Act: 0.0106. Min Act: -0.0006

Data Index: 382272 (C4 (Web Text))

Max Activating Token Index: 454