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

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

Neuron 16 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2910. Min Act: -0.0001

Data Index: 787336 (C4 (Web Text))

Max Activating Token Index: 94

Text #1

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2706. Min Act: -0.0001

Data Index: 60245 (C4 (Web Text))

Max Activating Token Index: 248

Text #2

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2799. Min Act: -0.0001

Data Index: 734767 (C4 (Web Text))

Max Activating Token Index: 88

Text #3

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2688. Min Act: -0.0001

Data Index: 467950 (C4 (Web Text))

Max Activating Token Index: 298

Text #4

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2632. Min Act: -0.0001

Data Index: 216415 (C4 (Web Text))

Max Activating Token Index: 252

Text #5

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2541. Min Act: -0.0001

Data Index: 586074 (C4 (Web Text))

Max Activating Token Index: 939

Text #6

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2406. Min Act: -0.0001

Data Index: 126728 (C4 (Web Text))

Max Activating Token Index: 262

Text #7

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2497. Min Act: -0.0001

Data Index: 472196 (C4 (Web Text))

Max Activating Token Index: 274

Text #8

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2384. Min Act: -0.0001

Data Index: 551568 (C4 (Web Text))

Max Activating Token Index: 298

Text #9

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2463. Min Act: -0.0001

Data Index: 390382 (C4 (Web Text))

Max Activating Token Index: 894

Text #10

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2521. Min Act: -0.0001

Data Index: 152703 (C4 (Web Text))

Max Activating Token Index: 197

Text #11

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2408. Min Act: -0.0001

Data Index: 684373 (C4 (Web Text))

Max Activating Token Index: 885

Text #12

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2338. Min Act: -0.0001

Data Index: 492458 (C4 (Web Text))

Max Activating Token Index: 399

Text #13

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2172. Min Act: -0.0001

Data Index: 1352727 (C4 (Web Text))

Max Activating Token Index: 568

Text #14

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2304. Min Act: -0.0001

Data Index: 588823 (C4 (Web Text))

Max Activating Token Index: 159

Text #15

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2128. Min Act: -0.0001

Data Index: 1106061 (C4 (Web Text))

Max Activating Token Index: 383

Text #16

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2190. Min Act: -0.0001

Data Index: 1102614 (C4 (Web Text))

Max Activating Token Index: 256

Text #17

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2093. Min Act: -0.0001

Data Index: 556887 (C4 (Web Text))

Max Activating Token Index: 243

Text #18

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.2074. Min Act: -0.0001

Data Index: 216582 (C4 (Web Text))

Max Activating Token Index: 815

Text #19

Max Range: 0.2910. Min Range: -0.2910

Max Act: 0.1996. Min Act: -0.0001

Data Index: 671804 (C4 (Web Text))

Max Activating Token Index: 133