Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 359 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0098. Min Act: -0.0006

Data Index: 1236769 (C4 (Web Text))

Max Activating Token Index: 728

Text #1

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0085. Min Act: -0.0006

Data Index: 61507 (C4 (Web Text))

Max Activating Token Index: 780

Text #2

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0076. Min Act: -0.0006

Data Index: 1377958 (Python Code)

Max Activating Token Index: 784

Text #3

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0072. Min Act: -0.0006

Data Index: 124680 (C4 (Web Text))

Max Activating Token Index: 490

Text #4

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0071. Min Act: -0.0006

Data Index: 622370 (C4 (Web Text))

Max Activating Token Index: 620

Text #5

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0071. Min Act: -0.0006

Data Index: 1352007 (C4 (Web Text))

Max Activating Token Index: 558

Text #6

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0070. Min Act: -0.0006

Data Index: 547664 (C4 (Web Text))

Max Activating Token Index: 147

Text #7

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0068. Min Act: -0.0005

Data Index: 196549 (C4 (Web Text))

Max Activating Token Index: 312

Text #8

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0068. Min Act: -0.0005

Data Index: 1366421 (Python Code)

Max Activating Token Index: 43

Text #9

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0067. Min Act: -0.0006

Data Index: 796943 (C4 (Web Text))

Max Activating Token Index: 869

Text #10

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0066. Min Act: -0.0006

Data Index: 1275976 (C4 (Web Text))

Max Activating Token Index: 645

Text #11

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0068. Min Act: -0.0006

Data Index: 254599 (C4 (Web Text))

Max Activating Token Index: 630

Text #12

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0006

Data Index: 398212 (C4 (Web Text))

Max Activating Token Index: 477

Text #13

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0006

Data Index: 929936 (C4 (Web Text))

Max Activating Token Index: 914

Text #14

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0006

Data Index: 1513883 (Python Code)

Max Activating Token Index: 126

Text #15

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0065. Min Act: -0.0006

Data Index: 602108 (C4 (Web Text))

Max Activating Token Index: 106

Text #16

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0006

Data Index: 628875 (C4 (Web Text))

Max Activating Token Index: 200

Text #17

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0065. Min Act: -0.0006

Data Index: 1357081 (C4 (Web Text))

Max Activating Token Index: 618

Text #18

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0005

Data Index: 405016 (C4 (Web Text))

Max Activating Token Index: 326

Text #19

Max Range: 0.0098. Min Range: -0.0098

Max Act: 0.0064. Min Act: -0.0006

Data Index: 548960 (C4 (Web Text))

Max Activating Token Index: 438