Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 2908 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.8470. Min Act: -0.0000

Data Index: 278757 (C4 (Web Text))

Max Activating Token Index: 982

Text #1

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.4430. Min Act: -0.0000

Data Index: 1469298 (Python Code)

Max Activating Token Index: 318

Text #2

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.3445. Min Act: -0.0000

Data Index: 523857 (C4 (Web Text))

Max Activating Token Index: 830

Text #3

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.3129. Min Act: -0.0000

Data Index: 819104 (C4 (Web Text))

Max Activating Token Index: 755

Text #4

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.0770. Min Act: -0.0000

Data Index: 1618526 (Python Code)

Max Activating Token Index: 979

Text #5

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.3079. Min Act: -0.0000

Data Index: 1384696 (Python Code)

Max Activating Token Index: 854

Text #6

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.2166. Min Act: -0.0000

Data Index: 1390477 (Python Code)

Max Activating Token Index: 284

Text #7

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.3214. Min Act: -0.0000

Data Index: 1055644 (C4 (Web Text))

Max Activating Token Index: 237

Text #8

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.9930. Min Act: -0.0000

Data Index: 1640307 (Python Code)

Max Activating Token Index: 690

Text #9

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.9630. Min Act: -0.0000

Data Index: 446888 (C4 (Web Text))

Max Activating Token Index: 265

Text #10

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.0320. Min Act: -0.0000

Data Index: 1575409 (Python Code)

Max Activating Token Index: 678

Text #11

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.9225. Min Act: -0.0000

Data Index: 268496 (C4 (Web Text))

Max Activating Token Index: 872

Text #12

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.1267. Min Act: -0.0000

Data Index: 584607 (C4 (Web Text))

Max Activating Token Index: 863

Text #13

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.0290. Min Act: -0.0000

Data Index: 738065 (C4 (Web Text))

Max Activating Token Index: 107

Text #14

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.9458. Min Act: -0.0000

Data Index: 1468052 (Python Code)

Max Activating Token Index: 594

Text #15

Max Range: 2.8470. Min Range: -2.8470

Max Act: 2.0073. Min Act: -0.0000

Data Index: 1589560 (Python Code)

Max Activating Token Index: 825

Text #16

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.7967. Min Act: -0.0000

Data Index: 1643301 (Python Code)

Max Activating Token Index: 135

Text #17

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.8871. Min Act: -0.0000

Data Index: 1654214 (Python Code)

Max Activating Token Index: 398

Text #18

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.9765. Min Act: -0.0000

Data Index: 1638557 (Python Code)

Max Activating Token Index: 307

Text #19

Max Range: 2.8470. Min Range: -2.8470

Max Act: 1.8962. Min Act: -0.0000

Data Index: 1345399 (C4 (Web Text))

Max Activating Token Index: 293