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

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

Neuron 2215 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.7871. Min Range: -2.7871

Max Act: 2.7871. Min Act: -0.0000

Data Index: 797866 (C4 (Web Text))

Max Activating Token Index: 813

Text #1

Max Range: 2.7871. Min Range: -2.7871

Max Act: 2.2273. Min Act: -0.0000

Data Index: 1135391 (C4 (Web Text))

Max Activating Token Index: 16

Text #2

Max Range: 2.7871. Min Range: -2.7871

Max Act: 2.2091. Min Act: -0.0000

Data Index: 555281 (C4 (Web Text))

Max Activating Token Index: 455

Text #3

Max Range: 2.7871. Min Range: -2.7871

Max Act: 2.1969. Min Act: -0.0000

Data Index: 730094 (C4 (Web Text))

Max Activating Token Index: 925

Text #4

Max Range: 2.7871. Min Range: -2.7871

Max Act: 2.0642. Min Act: -0.0000

Data Index: 507834 (C4 (Web Text))

Max Activating Token Index: 40

Text #5

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.9906. Min Act: -0.0000

Data Index: 491482 (C4 (Web Text))

Max Activating Token Index: 881

Text #6

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.9812. Min Act: -0.0000

Data Index: 1016782 (C4 (Web Text))

Max Activating Token Index: 380

Text #7

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.9593. Min Act: -0.0000

Data Index: 691697 (C4 (Web Text))

Max Activating Token Index: 119

Text #8

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.7905. Min Act: -0.0000

Data Index: 442992 (C4 (Web Text))

Max Activating Token Index: 349

Text #9

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.7772. Min Act: -0.0000

Data Index: 1184885 (C4 (Web Text))

Max Activating Token Index: 1007

Text #10

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.7518. Min Act: -0.0000

Data Index: 763460 (C4 (Web Text))

Max Activating Token Index: 20

Text #11

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6605. Min Act: -0.0000

Data Index: 1420096 (Python Code)

Max Activating Token Index: 148

Text #12

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.7347. Min Act: -0.0000

Data Index: 330709 (C4 (Web Text))

Max Activating Token Index: 328

Text #13

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6391. Min Act: -0.0000

Data Index: 486674 (C4 (Web Text))

Max Activating Token Index: 182

Text #14

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.7311. Min Act: -0.0000

Data Index: 1471419 (Python Code)

Max Activating Token Index: 802

Text #15

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6495. Min Act: -0.0000

Data Index: 5021 (C4 (Web Text))

Max Activating Token Index: 1005

Text #16

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6704. Min Act: -0.0000

Data Index: 1317791 (C4 (Web Text))

Max Activating Token Index: 612

Text #17

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6133. Min Act: -0.0000

Data Index: 361476 (C4 (Web Text))

Max Activating Token Index: 628

Text #18

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.6094. Min Act: -0.0000

Data Index: 947193 (C4 (Web Text))

Max Activating Token Index: 705

Text #19

Max Range: 2.7871. Min Range: -2.7871

Max Act: 1.4935. Min Act: -0.0000

Data Index: 292283 (C4 (Web Text))

Max Activating Token Index: 735