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

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

Neuron 3422 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: 0.0434. Min Range: -0.0434

Max Act: 0.0434. Min Act: -0.0002

Data Index: 1252981 (C4 (Web Text))

Max Activating Token Index: 405

Text #1

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0405. Min Act: -0.0002

Data Index: 1168105 (C4 (Web Text))

Max Activating Token Index: 246

Text #2

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0421. Min Act: -0.0002

Data Index: 156708 (C4 (Web Text))

Max Activating Token Index: 223

Text #3

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0404. Min Act: -0.0002

Data Index: 917139 (C4 (Web Text))

Max Activating Token Index: 222

Text #4

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0394. Min Act: -0.0001

Data Index: 1127379 (C4 (Web Text))

Max Activating Token Index: 247

Text #5

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0416. Min Act: -0.0002

Data Index: 910901 (C4 (Web Text))

Max Activating Token Index: 304

Text #6

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0384. Min Act: -0.0001

Data Index: 656741 (C4 (Web Text))

Max Activating Token Index: 888

Text #7

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0345. Min Act: -0.0002

Data Index: 1654363 (Python Code)

Max Activating Token Index: 707

Text #8

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0397. Min Act: -0.0002

Data Index: 442869 (C4 (Web Text))

Max Activating Token Index: 22

Text #9

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0396. Min Act: -0.0001

Data Index: 1374871 (Python Code)

Max Activating Token Index: 442

Text #10

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0402. Min Act: -0.0002

Data Index: 1325564 (C4 (Web Text))

Max Activating Token Index: 1018

Text #11

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0389. Min Act: -0.0002

Data Index: 1413947 (Python Code)

Max Activating Token Index: 403

Text #12

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0369. Min Act: -0.0001

Data Index: 1570189 (Python Code)

Max Activating Token Index: 109

Text #13

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0401. Min Act: -0.0002

Data Index: 1083506 (C4 (Web Text))

Max Activating Token Index: 938

Text #14

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0393. Min Act: -0.0002

Data Index: 393502 (C4 (Web Text))

Max Activating Token Index: 106

Text #15

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0384. Min Act: -0.0002

Data Index: 1189384 (C4 (Web Text))

Max Activating Token Index: 176

Text #16

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0367. Min Act: -0.0001

Data Index: 614476 (C4 (Web Text))

Max Activating Token Index: 222

Text #17

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0384. Min Act: -0.0001

Data Index: 1587417 (Python Code)

Max Activating Token Index: 943

Text #18

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0377. Min Act: -0.0001

Data Index: 963857 (C4 (Web Text))

Max Activating Token Index: 254

Text #19

Max Range: 0.0434. Min Range: -0.0434

Max Act: 0.0374. Min Act: -0.0001

Data Index: 52262 (C4 (Web Text))

Max Activating Token Index: 269