Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1105 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0507. Min Act: -0.0003

Data Index: 817202 (C4 (Web Text))

Max Activating Token Index: 145

Text #1

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0532. Min Act: -0.0003

Data Index: 370465 (C4 (Web Text))

Max Activating Token Index: 480

Text #2

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0531. Min Act: -0.0003

Data Index: 180305 (C4 (Web Text))

Max Activating Token Index: 358

Text #3

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0465. Min Act: -0.0003

Data Index: 748982 (C4 (Web Text))

Max Activating Token Index: 70

Text #4

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0479. Min Act: -0.0004

Data Index: 305729 (C4 (Web Text))

Max Activating Token Index: 415

Text #5

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0480. Min Act: -0.0003

Data Index: 6679 (C4 (Web Text))

Max Activating Token Index: 873

Text #6

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0455. Min Act: -0.0004

Data Index: 959921 (C4 (Web Text))

Max Activating Token Index: 281

Text #7

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0443. Min Act: -0.0004

Data Index: 937188 (C4 (Web Text))

Max Activating Token Index: 173

Text #8

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0442. Min Act: -0.0004

Data Index: 447720 (C4 (Web Text))

Max Activating Token Index: 32

Text #9

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0426. Min Act: -0.0003

Data Index: 1058402 (C4 (Web Text))

Max Activating Token Index: 588

Text #10

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0438. Min Act: -0.0003

Data Index: 683490 (C4 (Web Text))

Max Activating Token Index: 743

Text #11

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0431. Min Act: -0.0003

Data Index: 1007547 (C4 (Web Text))

Max Activating Token Index: 306

Text #12

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0414. Min Act: -0.0003

Data Index: 1115634 (C4 (Web Text))

Max Activating Token Index: 243

Text #13

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0416. Min Act: -0.0004

Data Index: 697413 (C4 (Web Text))

Max Activating Token Index: 600

Text #14

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0401. Min Act: -0.0004

Data Index: 92668 (C4 (Web Text))

Max Activating Token Index: 864

Text #15

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0410. Min Act: -0.0003

Data Index: 394650 (C4 (Web Text))

Max Activating Token Index: 124

Text #16

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0432. Min Act: -0.0003

Data Index: 198759 (C4 (Web Text))

Max Activating Token Index: 344

Text #17

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0418. Min Act: -0.0003

Data Index: 579926 (C4 (Web Text))

Max Activating Token Index: 782

Text #18

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0393. Min Act: -0.0003

Data Index: 152443 (C4 (Web Text))

Max Activating Token Index: 695

Text #19

Max Range: 0.0532. Min Range: -0.0532

Max Act: 0.0418. Min Act: -0.0003

Data Index: 342443 (C4 (Web Text))

Max Activating Token Index: 300