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

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

Neuron 5428 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.9296. Min Act: -0.0000

Data Index: 1466955 (Python Code)

Max Activating Token Index: 62

Text #1

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.9196. Min Act: -0.0000

Data Index: 1333833 (C4 (Web Text))

Max Activating Token Index: 800

Text #2

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7629. Min Act: -0.0000

Data Index: 1171438 (C4 (Web Text))

Max Activating Token Index: 805

Text #3

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7706. Min Act: -0.0000

Data Index: 235933 (C4 (Web Text))

Max Activating Token Index: 498

Text #4

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.8305. Min Act: -0.0000

Data Index: 58801 (C4 (Web Text))

Max Activating Token Index: 132

Text #5

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7663. Min Act: -0.0000

Data Index: 422308 (C4 (Web Text))

Max Activating Token Index: 991

Text #6

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7345. Min Act: -0.0000

Data Index: 1085670 (C4 (Web Text))

Max Activating Token Index: 841

Text #7

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7007. Min Act: -0.0000

Data Index: 4478 (C4 (Web Text))

Max Activating Token Index: 463

Text #8

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6769. Min Act: -0.0000

Data Index: 808466 (C4 (Web Text))

Max Activating Token Index: 401

Text #9

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6773. Min Act: -0.0000

Data Index: 923255 (C4 (Web Text))

Max Activating Token Index: 789

Text #10

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6805. Min Act: -0.0000

Data Index: 1545301 (Python Code)

Max Activating Token Index: 881

Text #11

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.7020. Min Act: -0.0000

Data Index: 162606 (C4 (Web Text))

Max Activating Token Index: 922

Text #12

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6548. Min Act: -0.0000

Data Index: 553946 (C4 (Web Text))

Max Activating Token Index: 647

Text #13

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6495. Min Act: -0.0000

Data Index: 1571966 (Python Code)

Max Activating Token Index: 630

Text #14

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6490. Min Act: -0.0000

Data Index: 11310 (C4 (Web Text))

Max Activating Token Index: 690

Text #15

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6127. Min Act: -0.0000

Data Index: 1367800 (Python Code)

Max Activating Token Index: 900

Text #16

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6739. Min Act: -0.0000

Data Index: 178829 (C4 (Web Text))

Max Activating Token Index: 677

Text #17

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6502. Min Act: -0.0000

Data Index: 677651 (C4 (Web Text))

Max Activating Token Index: 864

Text #18

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.5750. Min Act: -0.0000

Data Index: 1035105 (C4 (Web Text))

Max Activating Token Index: 744

Text #19

Max Range: 0.9296. Min Range: -0.9296

Max Act: 0.6724. Min Act: -0.0000

Data Index: 27456 (C4 (Web Text))

Max Activating Token Index: 294