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

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

Neuron 4418 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.5588. Min Act: -0.0000

Data Index: 1083540 (C4 (Web Text))

Max Activating Token Index: 50

Text #1

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.4377. Min Act: -0.0000

Data Index: 1199580 (C4 (Web Text))

Max Activating Token Index: 864

Text #2

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3679. Min Act: -0.0000

Data Index: 566100 (C4 (Web Text))

Max Activating Token Index: 263

Text #3

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3433. Min Act: -0.0000

Data Index: 452637 (C4 (Web Text))

Max Activating Token Index: 624

Text #4

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3517. Min Act: -0.0000

Data Index: 388405 (C4 (Web Text))

Max Activating Token Index: 506

Text #5

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3519. Min Act: -0.0000

Data Index: 163286 (C4 (Web Text))

Max Activating Token Index: 62

Text #6

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3457. Min Act: -0.0000

Data Index: 664966 (C4 (Web Text))

Max Activating Token Index: 754

Text #7

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3371. Min Act: -0.0000

Data Index: 434375 (C4 (Web Text))

Max Activating Token Index: 945

Text #8

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3470. Min Act: -0.0000

Data Index: 638255 (C4 (Web Text))

Max Activating Token Index: 190

Text #9

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3329. Min Act: -0.0000

Data Index: 768543 (C4 (Web Text))

Max Activating Token Index: 553

Text #10

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3201. Min Act: -0.0000

Data Index: 500730 (C4 (Web Text))

Max Activating Token Index: 258

Text #11

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3307. Min Act: -0.0000

Data Index: 1206095 (C4 (Web Text))

Max Activating Token Index: 372

Text #12

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3313. Min Act: -0.0000

Data Index: 952739 (C4 (Web Text))

Max Activating Token Index: 107

Text #13

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3267. Min Act: -0.0000

Data Index: 982150 (C4 (Web Text))

Max Activating Token Index: 388

Text #14

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3238. Min Act: -0.0000

Data Index: 409700 (C4 (Web Text))

Max Activating Token Index: 28

Text #15

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3245. Min Act: -0.0000

Data Index: 73170 (C4 (Web Text))

Max Activating Token Index: 432

Text #16

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3253. Min Act: -0.0000

Data Index: 730254 (C4 (Web Text))

Max Activating Token Index: 269

Text #17

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3198. Min Act: -0.0000

Data Index: 844293 (C4 (Web Text))

Max Activating Token Index: 554

Text #18

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3086. Min Act: -0.0000

Data Index: 635595 (C4 (Web Text))

Max Activating Token Index: 383

Text #19

Max Range: 0.5588. Min Range: -0.5588

Max Act: 0.3088. Min Act: -0.0000

Data Index: 371773 (C4 (Web Text))

Max Activating Token Index: 925