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

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

Neuron 4970 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.3423. Min Act: -0.0000

Data Index: 918807 (C4 (Web Text))

Max Activating Token Index: 168

Text #1

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.3158. Min Act: -0.0000

Data Index: 1588662 (Python Code)

Max Activating Token Index: 24

Text #2

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.2392. Min Act: -0.0000

Data Index: 1583370 (Python Code)

Max Activating Token Index: 20

Text #3

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.2867. Min Act: -0.0000

Data Index: 1369548 (Python Code)

Max Activating Token Index: 926

Text #4

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.2506. Min Act: -0.0000

Data Index: 1570175 (Python Code)

Max Activating Token Index: 552

Text #5

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.2037. Min Act: -0.0000

Data Index: 1397291 (Python Code)

Max Activating Token Index: 999

Text #6

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.2765. Min Act: -0.0000

Data Index: 1278330 (C4 (Web Text))

Max Activating Token Index: 321

Text #7

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.1650. Min Act: -0.0000

Data Index: 289336 (C4 (Web Text))

Max Activating Token Index: 565

Text #8

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.1638. Min Act: -0.0000

Data Index: 1078592 (C4 (Web Text))

Max Activating Token Index: 501

Text #9

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.1417. Min Act: -0.0000

Data Index: 247392 (C4 (Web Text))

Max Activating Token Index: 120

Text #10

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.1065. Min Act: -0.0000

Data Index: 748377 (C4 (Web Text))

Max Activating Token Index: 814

Text #11

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0950. Min Act: -0.0000

Data Index: 1505929 (Python Code)

Max Activating Token Index: 639

Text #12

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0732. Min Act: -0.0000

Data Index: 568761 (C4 (Web Text))

Max Activating Token Index: 978

Text #13

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0649. Min Act: -0.0000

Data Index: 1492936 (Python Code)

Max Activating Token Index: 961

Text #14

Max Range: 1.3423. Min Range: -1.3423

Max Act: 0.9824. Min Act: -0.0000

Data Index: 1386290 (Python Code)

Max Activating Token Index: 386

Text #15

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0496. Min Act: -0.0000

Data Index: 1398133 (Python Code)

Max Activating Token Index: 418

Text #16

Max Range: 1.3423. Min Range: -1.3423

Max Act: 0.9756. Min Act: -0.0000

Data Index: 635212 (C4 (Web Text))

Max Activating Token Index: 732

Text #17

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0357. Min Act: -0.0000

Data Index: 1413216 (Python Code)

Max Activating Token Index: 467

Text #18

Max Range: 1.3423. Min Range: -1.3423

Max Act: 0.9658. Min Act: -0.0000

Data Index: 1525951 (Python Code)

Max Activating Token Index: 640

Text #19

Max Range: 1.3423. Min Range: -1.3423

Max Act: 1.0230. Min Act: -0.0000

Data Index: 426527 (C4 (Web Text))

Max Activating Token Index: 619