Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 91 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0145. Min Act: -0.0001

Data Index: 398078 (C4 (Web Text))

Max Activating Token Index: 492

Text #1

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0131. Min Act: -0.0001

Data Index: 823412 (C4 (Web Text))

Max Activating Token Index: 312

Text #2

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0128. Min Act: -0.0001

Data Index: 88389 (C4 (Web Text))

Max Activating Token Index: 711

Text #3

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0123. Min Act: -0.0001

Data Index: 691521 (C4 (Web Text))

Max Activating Token Index: 507

Text #4

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0120. Min Act: -0.0001

Data Index: 877511 (C4 (Web Text))

Max Activating Token Index: 426

Text #5

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0116. Min Act: -0.0001

Data Index: 229177 (C4 (Web Text))

Max Activating Token Index: 650

Text #6

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0117. Min Act: -0.0001

Data Index: 283255 (C4 (Web Text))

Max Activating Token Index: 245

Text #7

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0118. Min Act: -0.0001

Data Index: 331783 (C4 (Web Text))

Max Activating Token Index: 415

Text #8

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0111. Min Act: -0.0001

Data Index: 228494 (C4 (Web Text))

Max Activating Token Index: 314

Text #9

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0113. Min Act: -0.0001

Data Index: 1131832 (C4 (Web Text))

Max Activating Token Index: 653

Text #10

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0114. Min Act: -0.0001

Data Index: 857484 (C4 (Web Text))

Max Activating Token Index: 504

Text #11

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0115. Min Act: -0.0001

Data Index: 849697 (C4 (Web Text))

Max Activating Token Index: 445

Text #12

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0112. Min Act: -0.0001

Data Index: 795284 (C4 (Web Text))

Max Activating Token Index: 448

Text #13

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0108. Min Act: -0.0001

Data Index: 558470 (C4 (Web Text))

Max Activating Token Index: 881

Text #14

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0123. Min Act: -0.0001

Data Index: 1485098 (Python Code)

Max Activating Token Index: 589

Text #15

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0117. Min Act: -0.0001

Data Index: 1175146 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0116. Min Act: -0.0001

Data Index: 184449 (C4 (Web Text))

Max Activating Token Index: 108

Text #17

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0116. Min Act: -0.0001

Data Index: 523356 (C4 (Web Text))

Max Activating Token Index: 920

Text #18

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0106. Min Act: -0.0001

Data Index: 1035365 (C4 (Web Text))

Max Activating Token Index: 638

Text #19

Max Range: 0.0145. Min Range: -0.0145

Max Act: 0.0108. Min Act: -0.0001

Data Index: 207192 (C4 (Web Text))

Max Activating Token Index: 801