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

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

Neuron 440 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.1058. Min Act: -0.0004

Data Index: 217133 (C4 (Web Text))

Max Activating Token Index: 954

Text #1

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0975. Min Act: -0.0003

Data Index: 246527 (C4 (Web Text))

Max Activating Token Index: 365

Text #2

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0815. Min Act: -0.0003

Data Index: 84134 (C4 (Web Text))

Max Activating Token Index: 233

Text #3

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0808. Min Act: -0.0003

Data Index: 522916 (C4 (Web Text))

Max Activating Token Index: 959

Text #4

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0740. Min Act: -0.0003

Data Index: 1342154 (C4 (Web Text))

Max Activating Token Index: 932

Text #5

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0761. Min Act: -0.0003

Data Index: 421729 (C4 (Web Text))

Max Activating Token Index: 95

Text #6

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0751. Min Act: -0.0004

Data Index: 518680 (C4 (Web Text))

Max Activating Token Index: 146

Text #7

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0735. Min Act: -0.0003

Data Index: 644064 (C4 (Web Text))

Max Activating Token Index: 395

Text #8

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0731. Min Act: -0.0002

Data Index: 402990 (C4 (Web Text))

Max Activating Token Index: 490

Text #9

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0740. Min Act: -0.0003

Data Index: 563921 (C4 (Web Text))

Max Activating Token Index: 606

Text #10

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0725. Min Act: -0.0003

Data Index: 435572 (C4 (Web Text))

Max Activating Token Index: 555

Text #11

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0765. Min Act: -0.0003

Data Index: 874069 (C4 (Web Text))

Max Activating Token Index: 506

Text #12

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0764. Min Act: -0.0003

Data Index: 128342 (C4 (Web Text))

Max Activating Token Index: 561

Text #13

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0711. Min Act: -0.0003

Data Index: 354050 (C4 (Web Text))

Max Activating Token Index: 524

Text #14

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0711. Min Act: -0.0003

Data Index: 218747 (C4 (Web Text))

Max Activating Token Index: 810

Text #15

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0718. Min Act: -0.0003

Data Index: 1029717 (C4 (Web Text))

Max Activating Token Index: 645

Text #16

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0711. Min Act: -0.0002

Data Index: 318233 (C4 (Web Text))

Max Activating Token Index: 983

Text #17

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0679. Min Act: -0.0003

Data Index: 203807 (C4 (Web Text))

Max Activating Token Index: 222

Text #18

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0697. Min Act: -0.0004

Data Index: 154663 (C4 (Web Text))

Max Activating Token Index: 697

Text #19

Max Range: 0.1058. Min Range: -0.1058

Max Act: 0.0684. Min Act: -0.0004

Data Index: 1198297 (C4 (Web Text))

Max Activating Token Index: 365