Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 754 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.5513. Min Act: -0.0002

Data Index: 126177 (C4 (Web Text))

Max Activating Token Index: 365

Text #1

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.3613. Min Act: -0.0002

Data Index: 966661 (C4 (Web Text))

Max Activating Token Index: 279

Text #2

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.5622. Min Act: -0.0001

Data Index: 17172 (C4 (Web Text))

Max Activating Token Index: 640

Text #3

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.5016. Min Act: -0.0001

Data Index: 371355 (C4 (Web Text))

Max Activating Token Index: 536

Text #4

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.4158. Min Act: -0.0001

Data Index: 1376541 (Python Code)

Max Activating Token Index: 845

Text #5

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.5590. Min Act: -0.0002

Data Index: 295725 (C4 (Web Text))

Max Activating Token Index: 891

Text #6

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.3630. Min Act: -0.0002

Data Index: 827413 (C4 (Web Text))

Max Activating Token Index: 959

Text #7

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2518. Min Act: -0.0002

Data Index: 1482484 (Python Code)

Max Activating Token Index: 751

Text #8

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.3188. Min Act: -0.0002

Data Index: 1286268 (C4 (Web Text))

Max Activating Token Index: 505

Text #9

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2461. Min Act: -0.0002

Data Index: 14952 (C4 (Web Text))

Max Activating Token Index: 230

Text #10

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2533. Min Act: -0.0002

Data Index: 244006 (C4 (Web Text))

Max Activating Token Index: 907

Text #11

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2135. Min Act: -0.0001

Data Index: 220610 (C4 (Web Text))

Max Activating Token Index: 619

Text #12

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2078. Min Act: -0.0002

Data Index: 983227 (C4 (Web Text))

Max Activating Token Index: 30

Text #13

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.1420. Min Act: -0.0002

Data Index: 168633 (C4 (Web Text))

Max Activating Token Index: 754

Text #14

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.3850. Min Act: -0.0001

Data Index: 844905 (C4 (Web Text))

Max Activating Token Index: 247

Text #15

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2068. Min Act: -0.0002

Data Index: 215712 (C4 (Web Text))

Max Activating Token Index: 312

Text #16

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.1398. Min Act: -0.0001

Data Index: 1458546 (Python Code)

Max Activating Token Index: 932

Text #17

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.2833. Min Act: -0.0002

Data Index: 394203 (C4 (Web Text))

Max Activating Token Index: 917

Text #18

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.1249. Min Act: -0.0002

Data Index: 129577 (C4 (Web Text))

Max Activating Token Index: 364

Text #19

Max Range: 1.5622. Min Range: -1.5622

Max Act: 1.1794. Min Act: -0.0001

Data Index: 519781 (C4 (Web Text))

Max Activating Token Index: 307