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

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

Neuron 3375 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.9441. Min Act: -0.0000

Data Index: 1446173 (Python Code)

Max Activating Token Index: 22

Text #1

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.5934. Min Act: -0.0000

Data Index: 675544 (C4 (Web Text))

Max Activating Token Index: 474

Text #2

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.7317. Min Act: -0.0000

Data Index: 91825 (C4 (Web Text))

Max Activating Token Index: 792

Text #3

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.8025. Min Act: -0.0000

Data Index: 1419979 (Python Code)

Max Activating Token Index: 57

Text #4

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.5285. Min Act: -0.0000

Data Index: 493369 (C4 (Web Text))

Max Activating Token Index: 363

Text #5

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.6260. Min Act: -0.0000

Data Index: 297605 (C4 (Web Text))

Max Activating Token Index: 770

Text #6

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.4555. Min Act: -0.0000

Data Index: 949891 (C4 (Web Text))

Max Activating Token Index: 850

Text #7

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.5141. Min Act: -0.0000

Data Index: 185287 (C4 (Web Text))

Max Activating Token Index: 568

Text #8

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.6355. Min Act: -0.0000

Data Index: 1597731 (Python Code)

Max Activating Token Index: 63

Text #9

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.4755. Min Act: -0.0000

Data Index: 1440473 (Python Code)

Max Activating Token Index: 588

Text #10

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.4700. Min Act: -0.0000

Data Index: 365346 (C4 (Web Text))

Max Activating Token Index: 752

Text #11

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.4755. Min Act: -0.0000

Data Index: 1371628 (Python Code)

Max Activating Token Index: 373

Text #12

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.2159. Min Act: -0.0000

Data Index: 1554735 (Python Code)

Max Activating Token Index: 265

Text #13

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.5639. Min Act: -0.0000

Data Index: 1469035 (Python Code)

Max Activating Token Index: 966

Text #14

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.3923. Min Act: -0.0000

Data Index: 1555521 (Python Code)

Max Activating Token Index: 71

Text #15

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.1461. Min Act: -0.0000

Data Index: 633971 (C4 (Web Text))

Max Activating Token Index: 576

Text #16

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.2607. Min Act: -0.0000

Data Index: 93700 (C4 (Web Text))

Max Activating Token Index: 34

Text #17

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.2537. Min Act: -0.0000

Data Index: 61566 (C4 (Web Text))

Max Activating Token Index: 585

Text #18

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.3732. Min Act: -0.0000

Data Index: 1340717 (C4 (Web Text))

Max Activating Token Index: 78

Text #19

Max Range: 2.9441. Min Range: -2.9441

Max Act: 2.3547. Min Act: -0.0000

Data Index: 1576444 (Python Code)

Max Activating Token Index: 583