Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 2902 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.7751. Min Act: -0.0001

Data Index: 443159 (C4 (Web Text))

Max Activating Token Index: 113

Text #1

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.7479. Min Act: -0.0001

Data Index: 538065 (C4 (Web Text))

Max Activating Token Index: 492

Text #2

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.6142. Min Act: -0.0001

Data Index: 1193052 (C4 (Web Text))

Max Activating Token Index: 481

Text #3

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.6173. Min Act: -0.0001

Data Index: 744090 (C4 (Web Text))

Max Activating Token Index: 910

Text #4

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.5343. Min Act: -0.0001

Data Index: 451831 (C4 (Web Text))

Max Activating Token Index: 520

Text #5

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.5129. Min Act: -0.0001

Data Index: 41248 (C4 (Web Text))

Max Activating Token Index: 648

Text #6

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.5111. Min Act: -0.0001

Data Index: 283177 (C4 (Web Text))

Max Activating Token Index: 629

Text #7

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.5117. Min Act: -0.0001

Data Index: 1601822 (Python Code)

Max Activating Token Index: 55

Text #8

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4487. Min Act: -0.0001

Data Index: 634174 (C4 (Web Text))

Max Activating Token Index: 813

Text #9

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4871. Min Act: -0.0001

Data Index: 518590 (C4 (Web Text))

Max Activating Token Index: 373

Text #10

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.5473. Min Act: -0.0001

Data Index: 250031 (C4 (Web Text))

Max Activating Token Index: 923

Text #11

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4065. Min Act: -0.0001

Data Index: 1461591 (Python Code)

Max Activating Token Index: 869

Text #12

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4877. Min Act: -0.0001

Data Index: 1491407 (Python Code)

Max Activating Token Index: 86

Text #13

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4485. Min Act: -0.0001

Data Index: 618584 (C4 (Web Text))

Max Activating Token Index: 521

Text #14

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4852. Min Act: -0.0001

Data Index: 1516915 (Python Code)

Max Activating Token Index: 306

Text #15

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.3930. Min Act: -0.0001

Data Index: 431608 (C4 (Web Text))

Max Activating Token Index: 57

Text #16

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4256. Min Act: -0.0001

Data Index: 765098 (C4 (Web Text))

Max Activating Token Index: 130

Text #17

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.4647. Min Act: -0.0001

Data Index: 1430470 (Python Code)

Max Activating Token Index: 693

Text #18

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.3686. Min Act: -0.0001

Data Index: 288281 (C4 (Web Text))

Max Activating Token Index: 195

Text #19

Max Range: 1.7751. Min Range: -1.7751

Max Act: 1.2497. Min Act: -0.0001

Data Index: 1526340 (Python Code)

Max Activating Token Index: 322