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

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

Neuron 6063 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: 4.1648. Min Range: -4.1648

Max Act: 4.1171. Min Act: -0.0000

Data Index: 620284 (C4 (Web Text))

Max Activating Token Index: 585

Text #1

Max Range: 4.1648. Min Range: -4.1648

Max Act: 4.1648. Min Act: -0.0000

Data Index: 712156 (C4 (Web Text))

Max Activating Token Index: 477

Text #2

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.6574. Min Act: -0.0000

Data Index: 681385 (C4 (Web Text))

Max Activating Token Index: 52

Text #3

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.6532. Min Act: -0.0000

Data Index: 1140808 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.4692. Min Act: -0.0000

Data Index: 145286 (C4 (Web Text))

Max Activating Token Index: 357

Text #5

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.5537. Min Act: -0.0000

Data Index: 7448 (C4 (Web Text))

Max Activating Token Index: 638

Text #6

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.5905. Min Act: -0.0000

Data Index: 1029373 (C4 (Web Text))

Max Activating Token Index: 776

Text #7

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.2069. Min Act: -0.0000

Data Index: 760759 (C4 (Web Text))

Max Activating Token Index: 910

Text #8

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.2435. Min Act: -0.0000

Data Index: 33181 (C4 (Web Text))

Max Activating Token Index: 1008

Text #9

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.1338. Min Act: -0.0000

Data Index: 978767 (C4 (Web Text))

Max Activating Token Index: 51

Text #10

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.2187. Min Act: -0.0000

Data Index: 1347531 (C4 (Web Text))

Max Activating Token Index: 442

Text #11

Max Range: 4.1648. Min Range: -4.1648

Max Act: 3.0385. Min Act: -0.0000

Data Index: 1174323 (C4 (Web Text))

Max Activating Token Index: 855

Text #12

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.7490. Min Act: -0.0000

Data Index: 1259247 (C4 (Web Text))

Max Activating Token Index: 191

Text #13

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.6973. Min Act: -0.0000

Data Index: 1361845 (Python Code)

Max Activating Token Index: 809

Text #14

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.9026. Min Act: -0.0000

Data Index: 87856 (C4 (Web Text))

Max Activating Token Index: 938

Text #15

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.7848. Min Act: -0.0000

Data Index: 597457 (C4 (Web Text))

Max Activating Token Index: 287

Text #16

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.7596. Min Act: -0.0000

Data Index: 707855 (C4 (Web Text))

Max Activating Token Index: 821

Text #17

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.7330. Min Act: -0.0000

Data Index: 325276 (C4 (Web Text))

Max Activating Token Index: 484

Text #18

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.8200. Min Act: -0.0000

Data Index: 979247 (C4 (Web Text))

Max Activating Token Index: 369

Text #19

Max Range: 4.1648. Min Range: -4.1648

Max Act: 2.6661. Min Act: -0.0000

Data Index: 1190705 (C4 (Web Text))

Max Activating Token Index: 40