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

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

Neuron 1460 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.7430. Min Act: -0.0000

Data Index: 272472 (C4 (Web Text))

Max Activating Token Index: 190

Text #1

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.7495. Min Act: -0.0000

Data Index: 14964 (C4 (Web Text))

Max Activating Token Index: 834

Text #2

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6674. Min Act: -0.0000

Data Index: 942145 (C4 (Web Text))

Max Activating Token Index: 786

Text #3

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6923. Min Act: -0.0000

Data Index: 254690 (C4 (Web Text))

Max Activating Token Index: 175

Text #4

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.7243. Min Act: -0.0000

Data Index: 552923 (C4 (Web Text))

Max Activating Token Index: 1015

Text #5

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.7101. Min Act: -0.0000

Data Index: 1030171 (C4 (Web Text))

Max Activating Token Index: 806

Text #6

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6482. Min Act: -0.0000

Data Index: 1053060 (C4 (Web Text))

Max Activating Token Index: 353

Text #7

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6896. Min Act: -0.0000

Data Index: 372933 (C4 (Web Text))

Max Activating Token Index: 108

Text #8

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6824. Min Act: -0.0000

Data Index: 1590710 (Python Code)

Max Activating Token Index: 276

Text #9

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6545. Min Act: -0.0000

Data Index: 952485 (C4 (Web Text))

Max Activating Token Index: 88

Text #10

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6522. Min Act: -0.0000

Data Index: 420364 (C4 (Web Text))

Max Activating Token Index: 757

Text #11

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6147. Min Act: -0.0000

Data Index: 204771 (C4 (Web Text))

Max Activating Token Index: 661

Text #12

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6284. Min Act: -0.0000

Data Index: 579938 (C4 (Web Text))

Max Activating Token Index: 362

Text #13

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6250. Min Act: -0.0000

Data Index: 83814 (C4 (Web Text))

Max Activating Token Index: 982

Text #14

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6584. Min Act: -0.0000

Data Index: 756302 (C4 (Web Text))

Max Activating Token Index: 135

Text #15

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6154. Min Act: -0.0000

Data Index: 229647 (C4 (Web Text))

Max Activating Token Index: 948

Text #16

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6025. Min Act: -0.0000

Data Index: 1258043 (C4 (Web Text))

Max Activating Token Index: 957

Text #17

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6328. Min Act: -0.0000

Data Index: 861866 (C4 (Web Text))

Max Activating Token Index: 560

Text #18

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.5702. Min Act: -0.0000

Data Index: 1194330 (C4 (Web Text))

Max Activating Token Index: 582

Text #19

Max Range: 0.7495. Min Range: -0.7495

Max Act: 0.6095. Min Act: -0.0000

Data Index: 389344 (C4 (Web Text))

Max Activating Token Index: 608