Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 215 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1852. Min Act: -0.0003

Data Index: 1443356 (Python Code)

Max Activating Token Index: 697

Text #1

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1686. Min Act: -0.0003

Data Index: 91333 (C4 (Web Text))

Max Activating Token Index: 265

Text #2

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1670. Min Act: -0.0003

Data Index: 238515 (C4 (Web Text))

Max Activating Token Index: 480

Text #3

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1659. Min Act: -0.0003

Data Index: 936598 (C4 (Web Text))

Max Activating Token Index: 639

Text #4

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1571. Min Act: -0.0003

Data Index: 533723 (C4 (Web Text))

Max Activating Token Index: 510

Text #5

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1590. Min Act: -0.0003

Data Index: 1528500 (Python Code)

Max Activating Token Index: 689

Text #6

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1570. Min Act: -0.0004

Data Index: 1183678 (C4 (Web Text))

Max Activating Token Index: 632

Text #7

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1491. Min Act: -0.0003

Data Index: 833023 (C4 (Web Text))

Max Activating Token Index: 641

Text #8

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1563. Min Act: -0.0003

Data Index: 1304316 (C4 (Web Text))

Max Activating Token Index: 51

Text #9

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1521. Min Act: -0.0003

Data Index: 1142609 (C4 (Web Text))

Max Activating Token Index: 230

Text #10

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1404. Min Act: -0.0004

Data Index: 1342743 (C4 (Web Text))

Max Activating Token Index: 949

Text #11

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1468. Min Act: -0.0003

Data Index: 1497173 (Python Code)

Max Activating Token Index: 409

Text #12

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1405. Min Act: -0.0003

Data Index: 1477137 (Python Code)

Max Activating Token Index: 999

Text #13

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1522. Min Act: -0.0003

Data Index: 563908 (C4 (Web Text))

Max Activating Token Index: 808

Text #14

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1325. Min Act: -0.0003

Data Index: 1626532 (Python Code)

Max Activating Token Index: 73

Text #15

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1400. Min Act: -0.0003

Data Index: 1505527 (Python Code)

Max Activating Token Index: 686

Text #16

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1397. Min Act: -0.0003

Data Index: 1462852 (Python Code)

Max Activating Token Index: 590

Text #17

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1368. Min Act: -0.0003

Data Index: 1268407 (C4 (Web Text))

Max Activating Token Index: 178

Text #18

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1351. Min Act: -0.0003

Data Index: 1351304 (C4 (Web Text))

Max Activating Token Index: 432

Text #19

Max Range: 0.1852. Min Range: -0.1852

Max Act: 0.1370. Min Act: -0.0004

Data Index: 1086234 (C4 (Web Text))

Max Activating Token Index: 490