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

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

Neuron 3707 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: 0.1723. Min Range: -0.1723

Max Act: 0.1723. Min Act: -0.0001

Data Index: 1404030 (Python Code)

Max Activating Token Index: 328

Text #1

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1636. Min Act: -0.0001

Data Index: 448081 (C4 (Web Text))

Max Activating Token Index: 195

Text #2

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1487. Min Act: -0.0001

Data Index: 115400 (C4 (Web Text))

Max Activating Token Index: 1019

Text #3

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1681. Min Act: -0.0001

Data Index: 1468724 (Python Code)

Max Activating Token Index: 325

Text #4

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1560. Min Act: -0.0001

Data Index: 1639133 (Python Code)

Max Activating Token Index: 262

Text #5

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1461. Min Act: -0.0001

Data Index: 1483527 (Python Code)

Max Activating Token Index: 409

Text #6

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1498. Min Act: -0.0001

Data Index: 1291597 (C4 (Web Text))

Max Activating Token Index: 153

Text #7

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1492. Min Act: -0.0001

Data Index: 1553183 (Python Code)

Max Activating Token Index: 458

Text #8

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1554. Min Act: -0.0001

Data Index: 955342 (C4 (Web Text))

Max Activating Token Index: 866

Text #9

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1479. Min Act: -0.0001

Data Index: 1606118 (Python Code)

Max Activating Token Index: 168

Text #10

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1513. Min Act: -0.0001

Data Index: 1449843 (Python Code)

Max Activating Token Index: 768

Text #11

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1451. Min Act: -0.0001

Data Index: 1643023 (Python Code)

Max Activating Token Index: 899

Text #12

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1549. Min Act: -0.0001

Data Index: 229498 (C4 (Web Text))

Max Activating Token Index: 787

Text #13

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1430. Min Act: -0.0001

Data Index: 1634282 (Python Code)

Max Activating Token Index: 105

Text #14

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1564. Min Act: -0.0001

Data Index: 1522266 (Python Code)

Max Activating Token Index: 809

Text #15

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1479. Min Act: -0.0001

Data Index: 1362520 (Python Code)

Max Activating Token Index: 915

Text #16

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1435. Min Act: -0.0001

Data Index: 364863 (C4 (Web Text))

Max Activating Token Index: 265

Text #17

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1419. Min Act: -0.0001

Data Index: 1230562 (C4 (Web Text))

Max Activating Token Index: 368

Text #18

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1359. Min Act: -0.0000

Data Index: 1025158 (C4 (Web Text))

Max Activating Token Index: 980

Text #19

Max Range: 0.1723. Min Range: -0.1723

Max Act: 0.1309. Min Act: -0.0000

Data Index: 609711 (C4 (Web Text))

Max Activating Token Index: 870