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

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

Neuron 36 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: 0.0526. Min Range: -0.0526

Max Act: 0.0522. Min Act: -0.0003

Data Index: 1233893 (C4 (Web Text))

Max Activating Token Index: 967

Text #1

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0526. Min Act: -0.0003

Data Index: 1602449 (Python Code)

Max Activating Token Index: 353

Text #2

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0519. Min Act: -0.0003

Data Index: 253660 (C4 (Web Text))

Max Activating Token Index: 955

Text #3

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0514. Min Act: -0.0003

Data Index: 854436 (C4 (Web Text))

Max Activating Token Index: 56

Text #4

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0508. Min Act: -0.0003

Data Index: 264665 (C4 (Web Text))

Max Activating Token Index: 747

Text #5

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0491. Min Act: -0.0003

Data Index: 1576023 (Python Code)

Max Activating Token Index: 800

Text #6

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0484. Min Act: -0.0002

Data Index: 1413847 (Python Code)

Max Activating Token Index: 984

Text #7

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0476. Min Act: -0.0003

Data Index: 1445678 (Python Code)

Max Activating Token Index: 983

Text #8

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0492. Min Act: 0.0001

Data Index: 5093 (C4 (Web Text))

Max Activating Token Index: 451

Text #9

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0471. Min Act: -0.0002

Data Index: 747614 (C4 (Web Text))

Max Activating Token Index: 467

Text #10

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0484. Min Act: -0.0003

Data Index: 1617348 (Python Code)

Max Activating Token Index: 1009

Text #11

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0462. Min Act: -0.0002

Data Index: 1299060 (C4 (Web Text))

Max Activating Token Index: 201

Text #12

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0472. Min Act: -0.0003

Data Index: 696531 (C4 (Web Text))

Max Activating Token Index: 1006

Text #13

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0453. Min Act: -0.0003

Data Index: 274679 (C4 (Web Text))

Max Activating Token Index: 977

Text #14

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0470. Min Act: -0.0002

Data Index: 264731 (C4 (Web Text))

Max Activating Token Index: 880

Text #15

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0449. Min Act: -0.0003

Data Index: 876308 (C4 (Web Text))

Max Activating Token Index: 988

Text #16

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0449. Min Act: -0.0003

Data Index: 1567154 (Python Code)

Max Activating Token Index: 361

Text #17

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0443. Min Act: -0.0002

Data Index: 1260162 (C4 (Web Text))

Max Activating Token Index: 852

Text #18

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0452. Min Act: -0.0002

Data Index: 1336452 (C4 (Web Text))

Max Activating Token Index: 834

Text #19

Max Range: 0.0526. Min Range: -0.0526

Max Act: 0.0448. Min Act: -0.0003

Data Index: 1441836 (Python Code)

Max Activating Token Index: 135