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

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

Neuron 120 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0103. Min Act: -0.0003

Data Index: 507393 (C4 (Web Text))

Max Activating Token Index: 878

Text #1

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0103. Min Act: -0.0003

Data Index: 1614841 (Python Code)

Max Activating Token Index: 74

Text #2

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0099. Min Act: -0.0002

Data Index: 1337828 (C4 (Web Text))

Max Activating Token Index: 615

Text #3

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0102. Min Act: -0.0003

Data Index: 1636575 (Python Code)

Max Activating Token Index: 56

Text #4

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0101. Min Act: -0.0002

Data Index: 1385480 (Python Code)

Max Activating Token Index: 381

Text #5

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0095. Min Act: -0.0002

Data Index: 1390787 (Python Code)

Max Activating Token Index: 962

Text #6

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0093. Min Act: -0.0002

Data Index: 107731 (C4 (Web Text))

Max Activating Token Index: 396

Text #7

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0089. Min Act: -0.0002

Data Index: 580386 (C4 (Web Text))

Max Activating Token Index: 351

Text #8

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0091. Min Act: -0.0003

Data Index: 288849 (C4 (Web Text))

Max Activating Token Index: 671

Text #9

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0089. Min Act: -0.0002

Data Index: 652953 (C4 (Web Text))

Max Activating Token Index: 27

Text #10

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0088. Min Act: -0.0003

Data Index: 1281643 (C4 (Web Text))

Max Activating Token Index: 629

Text #11

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0087. Min Act: -0.0003

Data Index: 1389331 (Python Code)

Max Activating Token Index: 948

Text #12

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0083. Min Act: -0.0003

Data Index: 1531218 (Python Code)

Max Activating Token Index: 660

Text #13

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0084. Min Act: -0.0003

Data Index: 704528 (C4 (Web Text))

Max Activating Token Index: 487

Text #14

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0088. Min Act: -0.0002

Data Index: 458163 (C4 (Web Text))

Max Activating Token Index: 670

Text #15

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0083. Min Act: -0.0002

Data Index: 65152 (C4 (Web Text))

Max Activating Token Index: 460

Text #16

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0085. Min Act: -0.0002

Data Index: 1629205 (Python Code)

Max Activating Token Index: 700

Text #17

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0087. Min Act: -0.0003

Data Index: 1366020 (Python Code)

Max Activating Token Index: 582

Text #18

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0083. Min Act: -0.0003

Data Index: 1280951 (C4 (Web Text))

Max Activating Token Index: 452

Text #19

Max Range: 0.0103. Min Range: -0.0103

Max Act: 0.0083. Min Act: -0.0003

Data Index: 1550595 (Python Code)

Max Activating Token Index: 174