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

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

Neuron 1091 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: 0.0677. Min Range: -0.0677

Max Act: 0.0556. Min Act: -0.0001

Data Index: 1036498 (C4 (Web Text))

Max Activating Token Index: 748

Text #1

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0677. Min Act: -0.0001

Data Index: 419617 (C4 (Web Text))

Max Activating Token Index: 914

Text #2

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0611. Min Act: -0.0001

Data Index: 1039035 (C4 (Web Text))

Max Activating Token Index: 568

Text #3

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0562. Min Act: -0.0001

Data Index: 311662 (C4 (Web Text))

Max Activating Token Index: 960

Text #4

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0511. Min Act: -0.0001

Data Index: 1444899 (Python Code)

Max Activating Token Index: 103

Text #5

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0528. Min Act: -0.0001

Data Index: 733607 (C4 (Web Text))

Max Activating Token Index: 539

Text #6

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0531. Min Act: -0.0001

Data Index: 1385531 (Python Code)

Max Activating Token Index: 676

Text #7

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0527. Min Act: -0.0001

Data Index: 106470 (C4 (Web Text))

Max Activating Token Index: 587

Text #8

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0458. Min Act: -0.0001

Data Index: 1505435 (Python Code)

Max Activating Token Index: 126

Text #9

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0518. Min Act: -0.0001

Data Index: 142868 (C4 (Web Text))

Max Activating Token Index: 485

Text #10

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0501. Min Act: -0.0001

Data Index: 1186120 (C4 (Web Text))

Max Activating Token Index: 627

Text #11

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0487. Min Act: -0.0001

Data Index: 838020 (C4 (Web Text))

Max Activating Token Index: 319

Text #12

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0488. Min Act: -0.0001

Data Index: 777050 (C4 (Web Text))

Max Activating Token Index: 438

Text #13

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0479. Min Act: -0.0001

Data Index: 843284 (C4 (Web Text))

Max Activating Token Index: 901

Text #14

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0474. Min Act: -0.0001

Data Index: 1598602 (Python Code)

Max Activating Token Index: 868

Text #15

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0472. Min Act: -0.0001

Data Index: 740612 (C4 (Web Text))

Max Activating Token Index: 803

Text #16

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0501. Min Act: -0.0001

Data Index: 1577701 (Python Code)

Max Activating Token Index: 655

Text #17

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0480. Min Act: -0.0001

Data Index: 1525333 (Python Code)

Max Activating Token Index: 890

Text #18

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0449. Min Act: -0.0001

Data Index: 964155 (C4 (Web Text))

Max Activating Token Index: 914

Text #19

Max Range: 0.0677. Min Range: -0.0677

Max Act: 0.0450. Min Act: -0.0001

Data Index: 81706 (C4 (Web Text))

Max Activating Token Index: 677