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

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

Neuron 573 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.7892. Min Act: -0.0000

Data Index: 1571200 (Python Code)

Max Activating Token Index: 901

Text #1

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.6321. Min Act: -0.0000

Data Index: 737857 (C4 (Web Text))

Max Activating Token Index: 914

Text #2

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.4253. Min Act: -0.0000

Data Index: 1435214 (Python Code)

Max Activating Token Index: 614

Text #3

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.4161. Min Act: -0.0000

Data Index: 1474698 (Python Code)

Max Activating Token Index: 892

Text #4

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.3972. Min Act: -0.0000

Data Index: 1642174 (Python Code)

Max Activating Token Index: 89

Text #5

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.4419. Min Act: -0.0000

Data Index: 1149193 (C4 (Web Text))

Max Activating Token Index: 771

Text #6

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.9818. Min Act: -0.0000

Data Index: 1569107 (Python Code)

Max Activating Token Index: 924

Text #7

Max Range: 3.7892. Min Range: -3.7892

Max Act: 3.0805. Min Act: -0.0000

Data Index: 1282054 (C4 (Web Text))

Max Activating Token Index: 165

Text #8

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7928. Min Act: -0.0000

Data Index: 772902 (C4 (Web Text))

Max Activating Token Index: 760

Text #9

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7868. Min Act: -0.0000

Data Index: 321180 (C4 (Web Text))

Max Activating Token Index: 760

Text #10

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7429. Min Act: -0.0000

Data Index: 70244 (C4 (Web Text))

Max Activating Token Index: 474

Text #11

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.6795. Min Act: -0.0000

Data Index: 802349 (C4 (Web Text))

Max Activating Token Index: 822

Text #12

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.8010. Min Act: -0.0000

Data Index: 1479251 (Python Code)

Max Activating Token Index: 605

Text #13

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.9514. Min Act: -0.0000

Data Index: 927824 (C4 (Web Text))

Max Activating Token Index: 358

Text #14

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7852. Min Act: -0.0000

Data Index: 1028091 (C4 (Web Text))

Max Activating Token Index: 475

Text #15

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.6084. Min Act: -0.0000

Data Index: 397101 (C4 (Web Text))

Max Activating Token Index: 868

Text #16

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7654. Min Act: -0.0000

Data Index: 347617 (C4 (Web Text))

Max Activating Token Index: 997

Text #17

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.5814. Min Act: -0.0000

Data Index: 284243 (C4 (Web Text))

Max Activating Token Index: 728

Text #18

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.7160. Min Act: -0.0000

Data Index: 1082838 (C4 (Web Text))

Max Activating Token Index: 227

Text #19

Max Range: 3.7892. Min Range: -3.7892

Max Act: 2.8571. Min Act: -0.0000

Data Index: 547957 (C4 (Web Text))

Max Activating Token Index: 398