Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 22 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2591. Min Act: -0.0002

Data Index: 917062 (C4 (Web Text))

Max Activating Token Index: 145

Text #1

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2595. Min Act: -0.0002

Data Index: 365180 (C4 (Web Text))

Max Activating Token Index: 337

Text #2

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2516. Min Act: -0.0002

Data Index: 1206490 (C4 (Web Text))

Max Activating Token Index: 341

Text #3

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2426. Min Act: -0.0001

Data Index: 1144190 (C4 (Web Text))

Max Activating Token Index: 630

Text #4

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2400. Min Act: -0.0002

Data Index: 1091656 (C4 (Web Text))

Max Activating Token Index: 292

Text #5

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2347. Min Act: -0.0002

Data Index: 1210113 (C4 (Web Text))

Max Activating Token Index: 624

Text #6

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2266. Min Act: -0.0002

Data Index: 1337198 (C4 (Web Text))

Max Activating Token Index: 535

Text #7

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2244. Min Act: -0.0002

Data Index: 354219 (C4 (Web Text))

Max Activating Token Index: 560

Text #8

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2258. Min Act: -0.0002

Data Index: 1118121 (C4 (Web Text))

Max Activating Token Index: 156

Text #9

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2226. Min Act: -0.0002

Data Index: 686729 (C4 (Web Text))

Max Activating Token Index: 357

Text #10

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2151. Min Act: -0.0002

Data Index: 1081804 (C4 (Web Text))

Max Activating Token Index: 228

Text #11

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2027. Min Act: -0.0002

Data Index: 417765 (C4 (Web Text))

Max Activating Token Index: 807

Text #12

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2135. Min Act: -0.0002

Data Index: 926453 (C4 (Web Text))

Max Activating Token Index: 521

Text #13

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.1997. Min Act: -0.0002

Data Index: 63727 (C4 (Web Text))

Max Activating Token Index: 654

Text #14

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.1990. Min Act: -0.0002

Data Index: 474311 (C4 (Web Text))

Max Activating Token Index: 905

Text #15

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2009. Min Act: -0.0002

Data Index: 1223417 (C4 (Web Text))

Max Activating Token Index: 523

Text #16

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2039. Min Act: -0.0002

Data Index: 164528 (C4 (Web Text))

Max Activating Token Index: 424

Text #17

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2035. Min Act: -0.0002

Data Index: 782604 (C4 (Web Text))

Max Activating Token Index: 547

Text #18

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.1930. Min Act: -0.0002

Data Index: 1152313 (C4 (Web Text))

Max Activating Token Index: 316

Text #19

Max Range: 0.2595. Min Range: -0.2595

Max Act: 0.2015. Min Act: -0.0002

Data Index: 102473 (C4 (Web Text))

Max Activating Token Index: 435