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

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

Neuron 768 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 6.9007. Min Range: -6.9007

Max Act: 6.9007. Min Act: -0.0000

Data Index: 797284 (C4 (Web Text))

Max Activating Token Index: 675

Text #1

Max Range: 6.9007. Min Range: -6.9007

Max Act: 5.2164. Min Act: -0.0000

Data Index: 620499 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.8684. Min Act: -0.0000

Data Index: 1277654 (C4 (Web Text))

Max Activating Token Index: 958

Text #3

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.5959. Min Act: -0.0000

Data Index: 393312 (C4 (Web Text))

Max Activating Token Index: 206

Text #4

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.7396. Min Act: -0.0000

Data Index: 783785 (C4 (Web Text))

Max Activating Token Index: 55

Text #5

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.1534. Min Act: -0.0000

Data Index: 1044857 (C4 (Web Text))

Max Activating Token Index: 990

Text #6

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.2540. Min Act: -0.0000

Data Index: 529196 (C4 (Web Text))

Max Activating Token Index: 685

Text #7

Max Range: 6.9007. Min Range: -6.9007

Max Act: 3.0704. Min Act: -0.0000

Data Index: 754464 (C4 (Web Text))

Max Activating Token Index: 503

Text #8

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.8783. Min Act: -0.0000

Data Index: 30164 (C4 (Web Text))

Max Activating Token Index: 115

Text #9

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.5437. Min Act: -0.0000

Data Index: 805170 (C4 (Web Text))

Max Activating Token Index: 119

Text #10

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.9172. Min Act: -0.0000

Data Index: 753996 (C4 (Web Text))

Max Activating Token Index: 455

Text #11

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.7260. Min Act: -0.0000

Data Index: 7749 (C4 (Web Text))

Max Activating Token Index: 129

Text #12

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.7901. Min Act: -0.0000

Data Index: 550866 (C4 (Web Text))

Max Activating Token Index: 538

Text #13

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.6099. Min Act: -0.0000

Data Index: 840747 (C4 (Web Text))

Max Activating Token Index: 730

Text #14

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.7022. Min Act: -0.0000

Data Index: 1652391 (Python Code)

Max Activating Token Index: 505

Text #15

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.5038. Min Act: -0.0000

Data Index: 1619244 (Python Code)

Max Activating Token Index: 381

Text #16

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.4996. Min Act: -0.0000

Data Index: 1381399 (Python Code)

Max Activating Token Index: 802

Text #17

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.3195. Min Act: -0.0000

Data Index: 1392216 (Python Code)

Max Activating Token Index: 503

Text #18

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.4476. Min Act: -0.0000

Data Index: 1590053 (Python Code)

Max Activating Token Index: 461

Text #19

Max Range: 6.9007. Min Range: -6.9007

Max Act: 2.1660. Min Act: -0.0000

Data Index: 453969 (C4 (Web Text))

Max Activating Token Index: 718