Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 656 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.5807. Min Act: -0.0000

Data Index: 62896 (C4 (Web Text))

Max Activating Token Index: 110

Text #1

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.5284. Min Act: -0.0000

Data Index: 904901 (C4 (Web Text))

Max Activating Token Index: 898

Text #2

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.6535. Min Act: -0.0000

Data Index: 504710 (C4 (Web Text))

Max Activating Token Index: 166

Text #3

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.4629. Min Act: -0.0000

Data Index: 1349305 (C4 (Web Text))

Max Activating Token Index: 477

Text #4

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.3071. Min Act: -0.0000

Data Index: 333034 (C4 (Web Text))

Max Activating Token Index: 25

Text #5

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.2397. Min Act: -0.0000

Data Index: 1069730 (C4 (Web Text))

Max Activating Token Index: 257

Text #6

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.2272. Min Act: -0.0000

Data Index: 106861 (C4 (Web Text))

Max Activating Token Index: 38

Text #7

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.3412. Min Act: -0.0000

Data Index: 159205 (C4 (Web Text))

Max Activating Token Index: 254

Text #8

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1962. Min Act: -0.0000

Data Index: 6878 (C4 (Web Text))

Max Activating Token Index: 677

Text #9

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.2960. Min Act: -0.0000

Data Index: 234409 (C4 (Web Text))

Max Activating Token Index: 439

Text #10

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1554. Min Act: -0.0000

Data Index: 1635032 (Python Code)

Max Activating Token Index: 56

Text #11

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1722. Min Act: -0.0000

Data Index: 1040649 (C4 (Web Text))

Max Activating Token Index: 190

Text #12

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.2812. Min Act: -0.0000

Data Index: 591109 (C4 (Web Text))

Max Activating Token Index: 293

Text #13

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1465. Min Act: -0.0000

Data Index: 1558425 (Python Code)

Max Activating Token Index: 781

Text #14

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1306. Min Act: -0.0000

Data Index: 367055 (C4 (Web Text))

Max Activating Token Index: 902

Text #15

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.1268. Min Act: -0.0000

Data Index: 1095907 (C4 (Web Text))

Max Activating Token Index: 778

Text #16

Max Range: 2.6535. Min Range: -2.6535

Max Act: 1.8670. Min Act: -0.0000

Data Index: 1626243 (Python Code)

Max Activating Token Index: 362

Text #17

Max Range: 2.6535. Min Range: -2.6535

Max Act: 1.9948. Min Act: -0.0000

Data Index: 815713 (C4 (Web Text))

Max Activating Token Index: 251

Text #18

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.0528. Min Act: -0.0000

Data Index: 1185269 (C4 (Web Text))

Max Activating Token Index: 213

Text #19

Max Range: 2.6535. Min Range: -2.6535

Max Act: 2.0516. Min Act: -0.0000

Data Index: 695197 (C4 (Web Text))

Max Activating Token Index: 561