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

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

Neuron 418 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0755. Min Act: -0.0000

Data Index: 917228 (C4 (Web Text))

Max Activating Token Index: 350

Text #1

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0480. Min Act: -0.0000

Data Index: 299275 (C4 (Web Text))

Max Activating Token Index: 429

Text #2

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0461. Min Act: -0.0000

Data Index: 1146829 (C4 (Web Text))

Max Activating Token Index: 670

Text #3

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0370. Min Act: -0.0000

Data Index: 1244939 (C4 (Web Text))

Max Activating Token Index: 225

Text #4

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0506. Min Act: -0.0000

Data Index: 1282487 (C4 (Web Text))

Max Activating Token Index: 66

Text #5

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0463. Min Act: -0.0000

Data Index: 417982 (C4 (Web Text))

Max Activating Token Index: 211

Text #6

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0419. Min Act: -0.0000

Data Index: 1191083 (C4 (Web Text))

Max Activating Token Index: 262

Text #7

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0449. Min Act: -0.0001

Data Index: 733030 (C4 (Web Text))

Max Activating Token Index: 896

Text #8

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0420. Min Act: -0.0001

Data Index: 66861 (C4 (Web Text))

Max Activating Token Index: 1003

Text #9

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0400. Min Act: -0.0000

Data Index: 318349 (C4 (Web Text))

Max Activating Token Index: 279

Text #10

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0402. Min Act: -0.0000

Data Index: 839086 (C4 (Web Text))

Max Activating Token Index: 763

Text #11

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0371. Min Act: -0.0000

Data Index: 1368803 (Python Code)

Max Activating Token Index: 582

Text #12

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0392. Min Act: -0.0000

Data Index: 1353588 (C4 (Web Text))

Max Activating Token Index: 145

Text #13

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0396. Min Act: -0.0000

Data Index: 1099626 (C4 (Web Text))

Max Activating Token Index: 560

Text #14

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0389. Min Act: -0.0000

Data Index: 396481 (C4 (Web Text))

Max Activating Token Index: 106

Text #15

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0351. Min Act: -0.0001

Data Index: 1058501 (C4 (Web Text))

Max Activating Token Index: 839

Text #16

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0347. Min Act: -0.0000

Data Index: 1060642 (C4 (Web Text))

Max Activating Token Index: 932

Text #17

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0373. Min Act: -0.0000

Data Index: 932339 (C4 (Web Text))

Max Activating Token Index: 58

Text #18

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0355. Min Act: -0.0000

Data Index: 718062 (C4 (Web Text))

Max Activating Token Index: 594

Text #19

Max Range: 0.0755. Min Range: -0.0755

Max Act: 0.0441. Min Act: -0.0000

Data Index: 1060767 (C4 (Web Text))

Max Activating Token Index: 300