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

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

Neuron 457 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1842. Min Act: -0.0000

Data Index: 250092 (C4 (Web Text))

Max Activating Token Index: 730

Text #1

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1679. Min Act: -0.0001

Data Index: 483035 (C4 (Web Text))

Max Activating Token Index: 844

Text #2

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1586. Min Act: -0.0000

Data Index: 377741 (C4 (Web Text))

Max Activating Token Index: 169

Text #3

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1549. Min Act: -0.0001

Data Index: 308848 (C4 (Web Text))

Max Activating Token Index: 388

Text #4

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1549. Min Act: -0.0001

Data Index: 1443897 (Python Code)

Max Activating Token Index: 321

Text #5

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1503. Min Act: -0.0000

Data Index: 758732 (C4 (Web Text))

Max Activating Token Index: 750

Text #6

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1495. Min Act: -0.0000

Data Index: 1329793 (C4 (Web Text))

Max Activating Token Index: 731

Text #7

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1433. Min Act: -0.0001

Data Index: 534135 (C4 (Web Text))

Max Activating Token Index: 154

Text #8

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1431. Min Act: -0.0001

Data Index: 1567690 (Python Code)

Max Activating Token Index: 607

Text #9

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1495. Min Act: -0.0001

Data Index: 1476694 (Python Code)

Max Activating Token Index: 55

Text #10

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1433. Min Act: -0.0001

Data Index: 1238677 (C4 (Web Text))

Max Activating Token Index: 367

Text #11

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1570. Min Act: -0.0001

Data Index: 283061 (C4 (Web Text))

Max Activating Token Index: 572

Text #12

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1335. Min Act: -0.0001

Data Index: 612647 (C4 (Web Text))

Max Activating Token Index: 568

Text #13

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1437. Min Act: -0.0001

Data Index: 358606 (C4 (Web Text))

Max Activating Token Index: 95

Text #14

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1376. Min Act: -0.0001

Data Index: 1281541 (C4 (Web Text))

Max Activating Token Index: 680

Text #15

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1316. Min Act: -0.0001

Data Index: 1480605 (Python Code)

Max Activating Token Index: 7

Text #16

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1339. Min Act: -0.0001

Data Index: 552568 (C4 (Web Text))

Max Activating Token Index: 102

Text #17

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1290. Min Act: -0.0001

Data Index: 1609381 (Python Code)

Max Activating Token Index: 6

Text #18

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1325. Min Act: -0.0001

Data Index: 487288 (C4 (Web Text))

Max Activating Token Index: 65

Text #19

Max Range: 0.1842. Min Range: -0.1842

Max Act: 0.1359. Min Act: -0.0001

Data Index: 1145573 (C4 (Web Text))

Max Activating Token Index: 1020