Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3033 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.2504. Min Range: -2.2504

Max Act: 2.2504. Min Act: -0.0000

Data Index: 104839 (C4 (Web Text))

Max Activating Token Index: 659

Text #1

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.9929. Min Act: -0.0000

Data Index: 1432995 (Python Code)

Max Activating Token Index: 999

Text #2

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.9484. Min Act: -0.0000

Data Index: 1520561 (Python Code)

Max Activating Token Index: 537

Text #3

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.8581. Min Act: -0.0000

Data Index: 1585823 (Python Code)

Max Activating Token Index: 348

Text #4

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.7855. Min Act: -0.0000

Data Index: 680628 (C4 (Web Text))

Max Activating Token Index: 35

Text #5

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.7811. Min Act: -0.0000

Data Index: 448631 (C4 (Web Text))

Max Activating Token Index: 955

Text #6

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.7971. Min Act: -0.0000

Data Index: 1564631 (Python Code)

Max Activating Token Index: 618

Text #7

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.6837. Min Act: -0.0000

Data Index: 1226822 (C4 (Web Text))

Max Activating Token Index: 599

Text #8

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4865. Min Act: -0.0000

Data Index: 1624495 (Python Code)

Max Activating Token Index: 731

Text #9

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.6127. Min Act: -0.0000

Data Index: 1532747 (Python Code)

Max Activating Token Index: 193

Text #10

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4741. Min Act: -0.0000

Data Index: 1452960 (Python Code)

Max Activating Token Index: 39

Text #11

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.5553. Min Act: -0.0000

Data Index: 1615498 (Python Code)

Max Activating Token Index: 8

Text #12

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.5378. Min Act: -0.0000

Data Index: 1436536 (Python Code)

Max Activating Token Index: 62

Text #13

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.5051. Min Act: -0.0000

Data Index: 861541 (C4 (Web Text))

Max Activating Token Index: 609

Text #14

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4115. Min Act: -0.0000

Data Index: 1619980 (Python Code)

Max Activating Token Index: 685

Text #15

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4831. Min Act: -0.0000

Data Index: 1570637 (Python Code)

Max Activating Token Index: 49

Text #16

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4790. Min Act: -0.0000

Data Index: 276160 (C4 (Web Text))

Max Activating Token Index: 84

Text #17

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.4751. Min Act: -0.0000

Data Index: 1423367 (Python Code)

Max Activating Token Index: 264

Text #18

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.3567. Min Act: -0.0000

Data Index: 1423627 (Python Code)

Max Activating Token Index: 150

Text #19

Max Range: 2.2504. Min Range: -2.2504

Max Act: 1.5153. Min Act: -0.0000

Data Index: 1550654 (Python Code)

Max Activating Token Index: 998