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

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

Neuron 3204 in Layer 11

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1000. Min Range: -1.1000

Max Act: 1.1000. Min Act: -0.0000

Data Index: 116525 (C4 (Web Text))

Max Activating Token Index: 562

Text #1

Max Range: 1.1000. Min Range: -1.1000

Max Act: 1.0717. Min Act: -0.0000

Data Index: 66211 (C4 (Web Text))

Max Activating Token Index: 124

Text #2

Max Range: 1.1000. Min Range: -1.1000

Max Act: 1.0867. Min Act: -0.0000

Data Index: 482654 (C4 (Web Text))

Max Activating Token Index: 703

Text #3

Max Range: 1.1000. Min Range: -1.1000

Max Act: 1.0176. Min Act: -0.0000

Data Index: 822108 (C4 (Web Text))

Max Activating Token Index: 985

Text #4

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.9603. Min Act: -0.0000

Data Index: 310390 (C4 (Web Text))

Max Activating Token Index: 771

Text #5

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8953. Min Act: -0.0000

Data Index: 301260 (C4 (Web Text))

Max Activating Token Index: 565

Text #6

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8922. Min Act: -0.0000

Data Index: 120981 (C4 (Web Text))

Max Activating Token Index: 244

Text #7

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8857. Min Act: -0.0000

Data Index: 46748 (C4 (Web Text))

Max Activating Token Index: 219

Text #8

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.9072. Min Act: -0.0000

Data Index: 113323 (C4 (Web Text))

Max Activating Token Index: 113

Text #9

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8903. Min Act: -0.0000

Data Index: 536064 (C4 (Web Text))

Max Activating Token Index: 897

Text #10

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8806. Min Act: -0.0000

Data Index: 655911 (C4 (Web Text))

Max Activating Token Index: 910

Text #11

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8007. Min Act: 0.0000

Data Index: 1465451 (Python Code)

Max Activating Token Index: 415

Text #12

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8982. Min Act: -0.0000

Data Index: 1460506 (Python Code)

Max Activating Token Index: 422

Text #13

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8330. Min Act: -0.0000

Data Index: 66562 (C4 (Web Text))

Max Activating Token Index: 443

Text #14

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8391. Min Act: -0.0000

Data Index: 1334498 (C4 (Web Text))

Max Activating Token Index: 701

Text #15

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8258. Min Act: -0.0000

Data Index: 231135 (C4 (Web Text))

Max Activating Token Index: 374

Text #16

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8197. Min Act: -0.0000

Data Index: 443502 (C4 (Web Text))

Max Activating Token Index: 598

Text #17

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.8836. Min Act: -0.0000

Data Index: 806408 (C4 (Web Text))

Max Activating Token Index: 972

Text #18

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.7789. Min Act: -0.0000

Data Index: 648739 (C4 (Web Text))

Max Activating Token Index: 692

Text #19

Max Range: 1.1000. Min Range: -1.1000

Max Act: 0.7373. Min Act: -0.0000

Data Index: 1036186 (C4 (Web Text))

Max Activating Token Index: 468