Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4267 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.9167. Min Act: -0.0000

Data Index: 308065 (C4 (Web Text))

Max Activating Token Index: 953

Text #1

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7190. Min Act: -0.0000

Data Index: 814658 (C4 (Web Text))

Max Activating Token Index: 755

Text #2

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7607. Min Act: -0.0000

Data Index: 656876 (C4 (Web Text))

Max Activating Token Index: 692

Text #3

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7092. Min Act: -0.0000

Data Index: 1190487 (C4 (Web Text))

Max Activating Token Index: 731

Text #4

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7231. Min Act: -0.0000

Data Index: 410996 (C4 (Web Text))

Max Activating Token Index: 1020

Text #5

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7071. Min Act: -0.0000

Data Index: 1482141 (Python Code)

Max Activating Token Index: 575

Text #6

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6801. Min Act: -0.0000

Data Index: 375483 (C4 (Web Text))

Max Activating Token Index: 198

Text #7

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6723. Min Act: -0.0000

Data Index: 1534225 (Python Code)

Max Activating Token Index: 157

Text #8

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.7110. Min Act: -0.0000

Data Index: 1100744 (C4 (Web Text))

Max Activating Token Index: 837

Text #9

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6773. Min Act: -0.0000

Data Index: 962532 (C4 (Web Text))

Max Activating Token Index: 264

Text #10

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6780. Min Act: -0.0000

Data Index: 743166 (C4 (Web Text))

Max Activating Token Index: 589

Text #11

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6783. Min Act: -0.0000

Data Index: 1181777 (C4 (Web Text))

Max Activating Token Index: 416

Text #12

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6829. Min Act: -0.0000

Data Index: 1045827 (C4 (Web Text))

Max Activating Token Index: 620

Text #13

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6695. Min Act: -0.0000

Data Index: 1459989 (Python Code)

Max Activating Token Index: 33

Text #14

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6643. Min Act: -0.0000

Data Index: 1043691 (C4 (Web Text))

Max Activating Token Index: 1006

Text #15

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6586. Min Act: -0.0000

Data Index: 4739 (C4 (Web Text))

Max Activating Token Index: 304

Text #16

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6154. Min Act: -0.0000

Data Index: 1622332 (Python Code)

Max Activating Token Index: 479

Text #17

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.5891. Min Act: -0.0000

Data Index: 1046261 (C4 (Web Text))

Max Activating Token Index: 1000

Text #18

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.5830. Min Act: -0.0000

Data Index: 11643 (C4 (Web Text))

Max Activating Token Index: 870

Text #19

Max Range: 0.9167. Min Range: -0.9167

Max Act: 0.6150. Min Act: -0.0000

Data Index: 989445 (C4 (Web Text))

Max Activating Token Index: 78