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

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

Neuron 3028 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.7000. Min Range: -4.7000

Max Act: 4.7000. Min Act: -0.0000

Data Index: 475514 (C4 (Web Text))

Max Activating Token Index: 915

Text #1

Max Range: 4.7000. Min Range: -4.7000

Max Act: 4.1350. Min Act: -0.0000

Data Index: 730369 (C4 (Web Text))

Max Activating Token Index: 1008

Text #2

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.7223. Min Act: -0.0000

Data Index: 870876 (C4 (Web Text))

Max Activating Token Index: 464

Text #3

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.8134. Min Act: -0.0000

Data Index: 1605611 (Python Code)

Max Activating Token Index: 974

Text #4

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.3871. Min Act: -0.0000

Data Index: 1075525 (C4 (Web Text))

Max Activating Token Index: 77

Text #5

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.2161. Min Act: -0.0000

Data Index: 818864 (C4 (Web Text))

Max Activating Token Index: 503

Text #6

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.4829. Min Act: -0.0000

Data Index: 1536485 (Python Code)

Max Activating Token Index: 907

Text #7

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.3266. Min Act: -0.0000

Data Index: 1550132 (Python Code)

Max Activating Token Index: 31

Text #8

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.1837. Min Act: -0.0000

Data Index: 509338 (C4 (Web Text))

Max Activating Token Index: 645

Text #9

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.2564. Min Act: -0.0000

Data Index: 379093 (C4 (Web Text))

Max Activating Token Index: 771

Text #10

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.9784. Min Act: -0.0000

Data Index: 1563216 (Python Code)

Max Activating Token Index: 310

Text #11

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.6767. Min Act: -0.0000

Data Index: 1126704 (C4 (Web Text))

Max Activating Token Index: 694

Text #12

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.5912. Min Act: -0.0000

Data Index: 13339 (C4 (Web Text))

Max Activating Token Index: 153

Text #13

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.8111. Min Act: -0.0001

Data Index: 952301 (C4 (Web Text))

Max Activating Token Index: 96

Text #14

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.6759. Min Act: -0.0000

Data Index: 727904 (C4 (Web Text))

Max Activating Token Index: 183

Text #15

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.9978. Min Act: -0.0000

Data Index: 1631237 (Python Code)

Max Activating Token Index: 996

Text #16

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.0918. Min Act: -0.0000

Data Index: 301037 (C4 (Web Text))

Max Activating Token Index: 421

Text #17

Max Range: 4.7000. Min Range: -4.7000

Max Act: 2.8650. Min Act: -0.0000

Data Index: 1242573 (C4 (Web Text))

Max Activating Token Index: 14

Text #18

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.1186. Min Act: -0.0000

Data Index: 238141 (C4 (Web Text))

Max Activating Token Index: 724

Text #19

Max Range: 4.7000. Min Range: -4.7000

Max Act: 3.1777. Min Act: -0.0000

Data Index: 1561165 (Python Code)

Max Activating Token Index: 988