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

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

Neuron 1506 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.0150. Min Range: -2.0150

Max Act: 2.0150. Min Act: -0.0000

Data Index: 1654927 (Python Code)

Max Activating Token Index: 295

Text #1

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.8009. Min Act: -0.0000

Data Index: 293400 (C4 (Web Text))

Max Activating Token Index: 671

Text #2

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.6991. Min Act: -0.0000

Data Index: 73854 (C4 (Web Text))

Max Activating Token Index: 637

Text #3

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.6359. Min Act: -0.0000

Data Index: 365625 (C4 (Web Text))

Max Activating Token Index: 55

Text #4

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.6360. Min Act: -0.0000

Data Index: 159776 (C4 (Web Text))

Max Activating Token Index: 1004

Text #5

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.5744. Min Act: -0.0000

Data Index: 1324549 (C4 (Web Text))

Max Activating Token Index: 817

Text #6

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.5415. Min Act: -0.0000

Data Index: 960539 (C4 (Web Text))

Max Activating Token Index: 485

Text #7

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.5400. Min Act: -0.0000

Data Index: 750755 (C4 (Web Text))

Max Activating Token Index: 318

Text #8

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.4999. Min Act: -0.0000

Data Index: 448964 (C4 (Web Text))

Max Activating Token Index: 771

Text #9

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.5547. Min Act: -0.0000

Data Index: 640127 (C4 (Web Text))

Max Activating Token Index: 939

Text #10

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.4165. Min Act: -0.0000

Data Index: 659321 (C4 (Web Text))

Max Activating Token Index: 1010

Text #11

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.3930. Min Act: -0.0000

Data Index: 325563 (C4 (Web Text))

Max Activating Token Index: 570

Text #12

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.3956. Min Act: -0.0000

Data Index: 282786 (C4 (Web Text))

Max Activating Token Index: 699

Text #13

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.4703. Min Act: -0.0000

Data Index: 252362 (C4 (Web Text))

Max Activating Token Index: 923

Text #14

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.2614. Min Act: -0.0000

Data Index: 1198137 (C4 (Web Text))

Max Activating Token Index: 80

Text #15

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.2995. Min Act: -0.0000

Data Index: 619657 (C4 (Web Text))

Max Activating Token Index: 331

Text #16

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.3655. Min Act: -0.0000

Data Index: 211209 (C4 (Web Text))

Max Activating Token Index: 569

Text #17

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.3140. Min Act: -0.0000

Data Index: 1454496 (Python Code)

Max Activating Token Index: 335

Text #18

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.2991. Min Act: -0.0000

Data Index: 502017 (C4 (Web Text))

Max Activating Token Index: 251

Text #19

Max Range: 2.0150. Min Range: -2.0150

Max Act: 1.3223. Min Act: -0.0000

Data Index: 1233255 (C4 (Web Text))

Max Activating Token Index: 791