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

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

Neuron 5009 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.2412. Min Range: -1.2412

Max Act: 1.2412. Min Act: -0.0000

Data Index: 1539991 (Python Code)

Max Activating Token Index: 335

Text #1

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.9065. Min Act: -0.0000

Data Index: 335003 (C4 (Web Text))

Max Activating Token Index: 292

Text #2

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8813. Min Act: -0.0000

Data Index: 1156795 (C4 (Web Text))

Max Activating Token Index: 996

Text #3

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8625. Min Act: -0.0000

Data Index: 252000 (C4 (Web Text))

Max Activating Token Index: 807

Text #4

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.9065. Min Act: -0.0000

Data Index: 367913 (C4 (Web Text))

Max Activating Token Index: 607

Text #5

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8430. Min Act: -0.0000

Data Index: 642112 (C4 (Web Text))

Max Activating Token Index: 578

Text #6

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8920. Min Act: -0.0000

Data Index: 413481 (C4 (Web Text))

Max Activating Token Index: 836

Text #7

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7792. Min Act: -0.0000

Data Index: 291565 (C4 (Web Text))

Max Activating Token Index: 82

Text #8

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8085. Min Act: -0.0000

Data Index: 498770 (C4 (Web Text))

Max Activating Token Index: 963

Text #9

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8069. Min Act: -0.0000

Data Index: 1123877 (C4 (Web Text))

Max Activating Token Index: 945

Text #10

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8266. Min Act: -0.0000

Data Index: 649896 (C4 (Web Text))

Max Activating Token Index: 446

Text #11

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8827. Min Act: -0.0000

Data Index: 175998 (C4 (Web Text))

Max Activating Token Index: 965

Text #12

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8391. Min Act: -0.0000

Data Index: 162461 (C4 (Web Text))

Max Activating Token Index: 526

Text #13

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8518. Min Act: -0.0000

Data Index: 1148158 (C4 (Web Text))

Max Activating Token Index: 76

Text #14

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7498. Min Act: -0.0000

Data Index: 1365527 (Python Code)

Max Activating Token Index: 385

Text #15

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7888. Min Act: -0.0000

Data Index: 287393 (C4 (Web Text))

Max Activating Token Index: 933

Text #16

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7862. Min Act: -0.0000

Data Index: 403335 (C4 (Web Text))

Max Activating Token Index: 570

Text #17

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.8067. Min Act: -0.0000

Data Index: 221252 (C4 (Web Text))

Max Activating Token Index: 222

Text #18

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7474. Min Act: -0.0000

Data Index: 653860 (C4 (Web Text))

Max Activating Token Index: 134

Text #19

Max Range: 1.2412. Min Range: -1.2412

Max Act: 0.7608. Min Act: -0.0000

Data Index: 826546 (C4 (Web Text))

Max Activating Token Index: 97