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

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

Neuron 3852 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.2258. Min Act: -0.0001

Data Index: 632590 (C4 (Web Text))

Max Activating Token Index: 294

Text #1

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1791. Min Act: -0.0001

Data Index: 1222923 (C4 (Web Text))

Max Activating Token Index: 135

Text #2

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1487. Min Act: -0.0001

Data Index: 1655044 (Python Code)

Max Activating Token Index: 148

Text #3

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1779. Min Act: -0.0001

Data Index: 448852 (C4 (Web Text))

Max Activating Token Index: 435

Text #4

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1903. Min Act: -0.0001

Data Index: 1492981 (Python Code)

Max Activating Token Index: 555

Text #5

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1591. Min Act: -0.0001

Data Index: 1604853 (Python Code)

Max Activating Token Index: 362

Text #6

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1493. Min Act: -0.0001

Data Index: 477148 (C4 (Web Text))

Max Activating Token Index: 90

Text #7

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1597. Min Act: -0.0001

Data Index: 1507616 (Python Code)

Max Activating Token Index: 40

Text #8

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1329. Min Act: -0.0001

Data Index: 375363 (C4 (Web Text))

Max Activating Token Index: 706

Text #9

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1758. Min Act: -0.0001

Data Index: 107445 (C4 (Web Text))

Max Activating Token Index: 86

Text #10

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1523. Min Act: -0.0001

Data Index: 11236 (C4 (Web Text))

Max Activating Token Index: 667

Text #11

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1708. Min Act: -0.0001

Data Index: 962971 (C4 (Web Text))

Max Activating Token Index: 710

Text #12

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1479. Min Act: -0.0001

Data Index: 166321 (C4 (Web Text))

Max Activating Token Index: 940

Text #13

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1465. Min Act: -0.0001

Data Index: 1505014 (Python Code)

Max Activating Token Index: 804

Text #14

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1516. Min Act: -0.0001

Data Index: 181187 (C4 (Web Text))

Max Activating Token Index: 678

Text #15

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1510. Min Act: -0.0001

Data Index: 1389430 (Python Code)

Max Activating Token Index: 826

Text #16

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1387. Min Act: -0.0001

Data Index: 627215 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1437. Min Act: -0.0001

Data Index: 1076578 (C4 (Web Text))

Max Activating Token Index: 195

Text #18

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1512. Min Act: -0.0001

Data Index: 624482 (C4 (Web Text))

Max Activating Token Index: 692

Text #19

Max Range: 0.2258. Min Range: -0.2258

Max Act: 0.1267. Min Act: -0.0001

Data Index: 1554640 (Python Code)

Max Activating Token Index: 289