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

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

Neuron 3123 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.3415. Min Act: -0.0000

Data Index: 553839 (C4 (Web Text))

Max Activating Token Index: 237

Text #1

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2546. Min Act: -0.0000

Data Index: 825243 (C4 (Web Text))

Max Activating Token Index: 658

Text #2

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.3209. Min Act: -0.0000

Data Index: 1142752 (C4 (Web Text))

Max Activating Token Index: 937

Text #3

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2714. Min Act: -0.0000

Data Index: 114062 (C4 (Web Text))

Max Activating Token Index: 192

Text #4

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2726. Min Act: -0.0000

Data Index: 506692 (C4 (Web Text))

Max Activating Token Index: 289

Text #5

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2538. Min Act: -0.0000

Data Index: 1359668 (C4 (Web Text))

Max Activating Token Index: 989

Text #6

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2417. Min Act: -0.0000

Data Index: 22579 (C4 (Web Text))

Max Activating Token Index: 19

Text #7

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2680. Min Act: -0.0000

Data Index: 824282 (C4 (Web Text))

Max Activating Token Index: 412

Text #8

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2425. Min Act: -0.0000

Data Index: 50461 (C4 (Web Text))

Max Activating Token Index: 376

Text #9

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2600. Min Act: -0.0000

Data Index: 1003926 (C4 (Web Text))

Max Activating Token Index: 111

Text #10

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2599. Min Act: -0.0000

Data Index: 190746 (C4 (Web Text))

Max Activating Token Index: 200

Text #11

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2335. Min Act: -0.0000

Data Index: 1080919 (C4 (Web Text))

Max Activating Token Index: 485

Text #12

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2592. Min Act: -0.0000

Data Index: 426665 (C4 (Web Text))

Max Activating Token Index: 996

Text #13

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2325. Min Act: -0.0000

Data Index: 965848 (C4 (Web Text))

Max Activating Token Index: 189

Text #14

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2340. Min Act: -0.0000

Data Index: 841553 (C4 (Web Text))

Max Activating Token Index: 488

Text #15

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2522. Min Act: -0.0000

Data Index: 29235 (C4 (Web Text))

Max Activating Token Index: 142

Text #16

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2461. Min Act: -0.0000

Data Index: 1509052 (Python Code)

Max Activating Token Index: 918

Text #17

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2659. Min Act: -0.0000

Data Index: 1295781 (C4 (Web Text))

Max Activating Token Index: 208

Text #18

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2426. Min Act: -0.0000

Data Index: 1150896 (C4 (Web Text))

Max Activating Token Index: 84

Text #19

Max Range: 0.3415. Min Range: -0.3415

Max Act: 0.2364. Min Act: -0.0000

Data Index: 191681 (C4 (Web Text))

Max Activating Token Index: 476