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

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

Neuron 4928 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.2537. Min Range: -4.2537

Max Act: 4.2537. Min Act: -0.0000

Data Index: 497940 (C4 (Web Text))

Max Activating Token Index: 842

Text #1

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.7734. Min Act: -0.0000

Data Index: 912641 (C4 (Web Text))

Max Activating Token Index: 879

Text #2

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.9935. Min Act: -0.0000

Data Index: 1010708 (C4 (Web Text))

Max Activating Token Index: 100

Text #3

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.4176. Min Act: -0.0000

Data Index: 123876 (C4 (Web Text))

Max Activating Token Index: 89

Text #4

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.4085. Min Act: -0.0000

Data Index: 1367942 (Python Code)

Max Activating Token Index: 112

Text #5

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.5590. Min Act: -0.0000

Data Index: 888772 (C4 (Web Text))

Max Activating Token Index: 585

Text #6

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.3818. Min Act: -0.0000

Data Index: 54436 (C4 (Web Text))

Max Activating Token Index: 90

Text #7

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.3394. Min Act: -0.0000

Data Index: 1310871 (C4 (Web Text))

Max Activating Token Index: 766

Text #8

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.4433. Min Act: -0.0000

Data Index: 119953 (C4 (Web Text))

Max Activating Token Index: 340

Text #9

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.2798. Min Act: -0.0000

Data Index: 375243 (C4 (Web Text))

Max Activating Token Index: 547

Text #10

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.3808. Min Act: -0.0000

Data Index: 688585 (C4 (Web Text))

Max Activating Token Index: 502

Text #11

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.3689. Min Act: -0.0000

Data Index: 203370 (C4 (Web Text))

Max Activating Token Index: 357

Text #12

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.2193. Min Act: -0.0000

Data Index: 787184 (C4 (Web Text))

Max Activating Token Index: 463

Text #13

Max Range: 4.2537. Min Range: -4.2537

Max Act: 2.9899. Min Act: -0.0000

Data Index: 381055 (C4 (Web Text))

Max Activating Token Index: 111

Text #14

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.1084. Min Act: -0.0000

Data Index: 198218 (C4 (Web Text))

Max Activating Token Index: 638

Text #15

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.2291. Min Act: -0.0000

Data Index: 532774 (C4 (Web Text))

Max Activating Token Index: 437

Text #16

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.0891. Min Act: -0.0000

Data Index: 540683 (C4 (Web Text))

Max Activating Token Index: 338

Text #17

Max Range: 4.2537. Min Range: -4.2537

Max Act: 2.9505. Min Act: -0.0000

Data Index: 1287711 (C4 (Web Text))

Max Activating Token Index: 137

Text #18

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.2267. Min Act: -0.0000

Data Index: 1642158 (Python Code)

Max Activating Token Index: 628

Text #19

Max Range: 4.2537. Min Range: -4.2537

Max Act: 3.2031. Min Act: -0.0000

Data Index: 1249027 (C4 (Web Text))

Max Activating Token Index: 1006