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

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

Neuron 30 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1872. Min Act: -0.0001

Data Index: 445622 (C4 (Web Text))

Max Activating Token Index: 913

Text #1

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1760. Min Act: -0.0001

Data Index: 1051252 (C4 (Web Text))

Max Activating Token Index: 16

Text #2

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1443. Min Act: -0.0001

Data Index: 305689 (C4 (Web Text))

Max Activating Token Index: 962

Text #3

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1466. Min Act: -0.0001

Data Index: 456533 (C4 (Web Text))

Max Activating Token Index: 386

Text #4

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1397. Min Act: -0.0001

Data Index: 136951 (C4 (Web Text))

Max Activating Token Index: 277

Text #5

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1576. Min Act: -0.0001

Data Index: 327592 (C4 (Web Text))

Max Activating Token Index: 840

Text #6

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1474. Min Act: -0.0001

Data Index: 213805 (C4 (Web Text))

Max Activating Token Index: 612

Text #7

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1464. Min Act: -0.0001

Data Index: 415403 (C4 (Web Text))

Max Activating Token Index: 334

Text #8

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1339. Min Act: -0.0001

Data Index: 880753 (C4 (Web Text))

Max Activating Token Index: 308

Text #9

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1413. Min Act: -0.0001

Data Index: 1476534 (Python Code)

Max Activating Token Index: 152

Text #10

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1386. Min Act: -0.0001

Data Index: 1142222 (C4 (Web Text))

Max Activating Token Index: 778

Text #11

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1330. Min Act: -0.0001

Data Index: 86885 (C4 (Web Text))

Max Activating Token Index: 601

Text #12

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1396. Min Act: -0.0001

Data Index: 428756 (C4 (Web Text))

Max Activating Token Index: 295

Text #13

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1495. Min Act: -0.0001

Data Index: 20316 (C4 (Web Text))

Max Activating Token Index: 576

Text #14

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1355. Min Act: -0.0001

Data Index: 857836 (C4 (Web Text))

Max Activating Token Index: 978

Text #15

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1447. Min Act: -0.0001

Data Index: 408528 (C4 (Web Text))

Max Activating Token Index: 143

Text #16

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1482. Min Act: -0.0001

Data Index: 704170 (C4 (Web Text))

Max Activating Token Index: 51

Text #17

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1452. Min Act: -0.0001

Data Index: 700051 (C4 (Web Text))

Max Activating Token Index: 700

Text #18

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1580. Min Act: -0.0001

Data Index: 172201 (C4 (Web Text))

Max Activating Token Index: 304

Text #19

Max Range: 0.1872. Min Range: -0.1872

Max Act: 0.1401. Min Act: -0.0001

Data Index: 1471331 (Python Code)

Max Activating Token Index: 41