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

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

Neuron 498 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.8409. Min Act: -0.0000

Data Index: 17661 (C4 (Web Text))

Max Activating Token Index: 889

Text #1

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.3460. Min Act: -0.0000

Data Index: 1521291 (Python Code)

Max Activating Token Index: 394

Text #2

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.5366. Min Act: -0.0000

Data Index: 1386542 (Python Code)

Max Activating Token Index: 210

Text #3

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1603. Min Act: -0.0000

Data Index: 655566 (C4 (Web Text))

Max Activating Token Index: 759

Text #4

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2975. Min Act: -0.0000

Data Index: 1536448 (Python Code)

Max Activating Token Index: 66

Text #5

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.7950. Min Act: -0.0000

Data Index: 1553893 (Python Code)

Max Activating Token Index: 235

Text #6

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2109. Min Act: -0.0000

Data Index: 1443488 (Python Code)

Max Activating Token Index: 743

Text #7

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2763. Min Act: -0.0000

Data Index: 1365465 (Python Code)

Max Activating Token Index: 935

Text #8

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1080. Min Act: -0.0000

Data Index: 824581 (C4 (Web Text))

Max Activating Token Index: 575

Text #9

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.4095. Min Act: -0.0000

Data Index: 1578640 (Python Code)

Max Activating Token Index: 929

Text #10

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1149. Min Act: -0.0000

Data Index: 846120 (C4 (Web Text))

Max Activating Token Index: 406

Text #11

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2587. Min Act: -0.0000

Data Index: 940942 (C4 (Web Text))

Max Activating Token Index: 9

Text #12

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1138. Min Act: -0.0000

Data Index: 299016 (C4 (Web Text))

Max Activating Token Index: 1009

Text #13

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2397. Min Act: -0.0000

Data Index: 476070 (C4 (Web Text))

Max Activating Token Index: 9

Text #14

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.2355. Min Act: -0.0000

Data Index: 709156 (C4 (Web Text))

Max Activating Token Index: 923

Text #15

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.3711. Min Act: -0.0000

Data Index: 1595109 (Python Code)

Max Activating Token Index: 677

Text #16

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1817. Min Act: -0.0000

Data Index: 509944 (C4 (Web Text))

Max Activating Token Index: 590

Text #17

Max Range: 3.8409. Min Range: -3.8409

Max Act: 2.8107. Min Act: -0.0000

Data Index: 1394398 (Python Code)

Max Activating Token Index: 211

Text #18

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1615. Min Act: -0.0000

Data Index: 1448520 (Python Code)

Max Activating Token Index: 867

Text #19

Max Range: 3.8409. Min Range: -3.8409

Max Act: 3.1541. Min Act: -0.0000

Data Index: 1451740 (Python Code)

Max Activating Token Index: 17