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

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

Neuron 13 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0670. Min Act: -0.0006

Data Index: 497636 (C4 (Web Text))

Max Activating Token Index: 557

Text #1

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0643. Min Act: -0.0006

Data Index: 1124768 (C4 (Web Text))

Max Activating Token Index: 141

Text #2

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0597. Min Act: -0.0006

Data Index: 938176 (C4 (Web Text))

Max Activating Token Index: 180

Text #3

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0588. Min Act: -0.0006

Data Index: 22037 (C4 (Web Text))

Max Activating Token Index: 114

Text #4

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0583. Min Act: -0.0006

Data Index: 288579 (C4 (Web Text))

Max Activating Token Index: 248

Text #5

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0608. Min Act: -0.0006

Data Index: 379495 (C4 (Web Text))

Max Activating Token Index: 954

Text #6

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0566. Min Act: -0.0006

Data Index: 781497 (C4 (Web Text))

Max Activating Token Index: 138

Text #7

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0521. Min Act: -0.0006

Data Index: 451074 (C4 (Web Text))

Max Activating Token Index: 453

Text #8

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0543. Min Act: -0.0006

Data Index: 22350 (C4 (Web Text))

Max Activating Token Index: 496

Text #9

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0552. Min Act: -0.0006

Data Index: 478029 (C4 (Web Text))

Max Activating Token Index: 110

Text #10

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0549. Min Act: -0.0006

Data Index: 849818 (C4 (Web Text))

Max Activating Token Index: 115

Text #11

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0537. Min Act: -0.0006

Data Index: 159711 (C4 (Web Text))

Max Activating Token Index: 550

Text #12

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0530. Min Act: -0.0006

Data Index: 714638 (C4 (Web Text))

Max Activating Token Index: 731

Text #13

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0519. Min Act: -0.0006

Data Index: 237078 (C4 (Web Text))

Max Activating Token Index: 506

Text #14

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0527. Min Act: -0.0006

Data Index: 304389 (C4 (Web Text))

Max Activating Token Index: 859

Text #15

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0531. Min Act: -0.0006

Data Index: 14437 (C4 (Web Text))

Max Activating Token Index: 150

Text #16

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0540. Min Act: -0.0006

Data Index: 1266670 (C4 (Web Text))

Max Activating Token Index: 128

Text #17

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0549. Min Act: -0.0006

Data Index: 528413 (C4 (Web Text))

Max Activating Token Index: 579

Text #18

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0513. Min Act: -0.0006

Data Index: 68443 (C4 (Web Text))

Max Activating Token Index: 227

Text #19

Max Range: 0.0670. Min Range: -0.0670

Max Act: 0.0533. Min Act: -0.0006

Data Index: 1307446 (C4 (Web Text))

Max Activating Token Index: 403