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

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

Neuron 4995 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.7673. Min Act: -0.0000

Data Index: 1112567 (C4 (Web Text))

Max Activating Token Index: 980

Text #1

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.6871. Min Act: -0.0000

Data Index: 960810 (C4 (Web Text))

Max Activating Token Index: 477

Text #2

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.2877. Min Act: -0.0000

Data Index: 111083 (C4 (Web Text))

Max Activating Token Index: 905

Text #3

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.3107. Min Act: -0.0000

Data Index: 1206976 (C4 (Web Text))

Max Activating Token Index: 159

Text #4

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.3112. Min Act: -0.0000

Data Index: 889349 (C4 (Web Text))

Max Activating Token Index: 426

Text #5

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.1654. Min Act: -0.0000

Data Index: 1061242 (C4 (Web Text))

Max Activating Token Index: 637

Text #6

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0910. Min Act: -0.0000

Data Index: 42039 (C4 (Web Text))

Max Activating Token Index: 577

Text #7

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0094. Min Act: -0.0000

Data Index: 600143 (C4 (Web Text))

Max Activating Token Index: 639

Text #8

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.3448. Min Act: -0.0000

Data Index: 1168482 (C4 (Web Text))

Max Activating Token Index: 290

Text #9

Max Range: 2.7673. Min Range: -2.7673

Max Act: 1.9857. Min Act: -0.0000

Data Index: 1113369 (C4 (Web Text))

Max Activating Token Index: 68

Text #10

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0400. Min Act: -0.0000

Data Index: 489117 (C4 (Web Text))

Max Activating Token Index: 929

Text #11

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0333. Min Act: -0.0000

Data Index: 421397 (C4 (Web Text))

Max Activating Token Index: 502

Text #12

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0606. Min Act: -0.0000

Data Index: 379443 (C4 (Web Text))

Max Activating Token Index: 795

Text #13

Max Range: 2.7673. Min Range: -2.7673

Max Act: 1.7611. Min Act: -0.0000

Data Index: 458677 (C4 (Web Text))

Max Activating Token Index: 245

Text #14

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.0347. Min Act: -0.0000

Data Index: 915886 (C4 (Web Text))

Max Activating Token Index: 633

Text #15

Max Range: 2.7673. Min Range: -2.7673

Max Act: 1.8793. Min Act: -0.0000

Data Index: 1253987 (C4 (Web Text))

Max Activating Token Index: 326

Text #16

Max Range: 2.7673. Min Range: -2.7673

Max Act: 1.9638. Min Act: -0.0000

Data Index: 395860 (C4 (Web Text))

Max Activating Token Index: 496

Text #17

Max Range: 2.7673. Min Range: -2.7673

Max Act: 1.8423. Min Act: -0.0000

Data Index: 490950 (C4 (Web Text))

Max Activating Token Index: 622

Text #18

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.1036. Min Act: -0.0000

Data Index: 1453073 (Python Code)

Max Activating Token Index: 337

Text #19

Max Range: 2.7673. Min Range: -2.7673

Max Act: 2.1540. Min Act: -0.0000

Data Index: 488761 (C4 (Web Text))

Max Activating Token Index: 374