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

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

Neuron 465 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: 0.0744. Min Range: -0.0744

Max Act: 0.0744. Min Act: -0.0003

Data Index: 1120504 (C4 (Web Text))

Max Activating Token Index: 532

Text #1

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0615. Min Act: -0.0003

Data Index: 1201558 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0606. Min Act: -0.0003

Data Index: 822778 (C4 (Web Text))

Max Activating Token Index: 366

Text #3

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0561. Min Act: -0.0003

Data Index: 1249499 (C4 (Web Text))

Max Activating Token Index: 526

Text #4

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0581. Min Act: -0.0002

Data Index: 992789 (C4 (Web Text))

Max Activating Token Index: 532

Text #5

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0540. Min Act: -0.0003

Data Index: 1094205 (C4 (Web Text))

Max Activating Token Index: 831

Text #6

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0551. Min Act: -0.0004

Data Index: 1139848 (C4 (Web Text))

Max Activating Token Index: 693

Text #7

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0498. Min Act: -0.0003

Data Index: 957370 (C4 (Web Text))

Max Activating Token Index: 942

Text #8

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0473. Min Act: -0.0003

Data Index: 548815 (C4 (Web Text))

Max Activating Token Index: 663

Text #9

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0466. Min Act: -0.0003

Data Index: 1532259 (Python Code)

Max Activating Token Index: 986

Text #10

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0504. Min Act: -0.0003

Data Index: 1576018 (Python Code)

Max Activating Token Index: 1017

Text #11

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0484. Min Act: -0.0003

Data Index: 745808 (C4 (Web Text))

Max Activating Token Index: 357

Text #12

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0457. Min Act: -0.0003

Data Index: 795528 (C4 (Web Text))

Max Activating Token Index: 454

Text #13

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0493. Min Act: -0.0003

Data Index: 1523388 (Python Code)

Max Activating Token Index: 803

Text #14

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0444. Min Act: -0.0004

Data Index: 155407 (C4 (Web Text))

Max Activating Token Index: 420

Text #15

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0467. Min Act: -0.0003

Data Index: 142758 (C4 (Web Text))

Max Activating Token Index: 785

Text #16

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0428. Min Act: -0.0003

Data Index: 1485177 (Python Code)

Max Activating Token Index: 1003

Text #17

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0458. Min Act: -0.0003

Data Index: 47295 (C4 (Web Text))

Max Activating Token Index: 297

Text #18

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0460. Min Act: -0.0003

Data Index: 353935 (C4 (Web Text))

Max Activating Token Index: 656

Text #19

Max Range: 0.0744. Min Range: -0.0744

Max Act: 0.0472. Min Act: -0.0003

Data Index: 1149820 (C4 (Web Text))

Max Activating Token Index: 733