Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 3579 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0582. Min Act: -0.0002

Data Index: 75411 (C4 (Web Text))

Max Activating Token Index: 162

Text #1

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0660. Min Act: -0.0002

Data Index: 190446 (C4 (Web Text))

Max Activating Token Index: 205

Text #2

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0559. Min Act: -0.0002

Data Index: 722977 (C4 (Web Text))

Max Activating Token Index: 9

Text #3

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0554. Min Act: -0.0002

Data Index: 364809 (C4 (Web Text))

Max Activating Token Index: 106

Text #4

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0504. Min Act: -0.0002

Data Index: 968796 (C4 (Web Text))

Max Activating Token Index: 573

Text #5

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0540. Min Act: -0.0002

Data Index: 502639 (C4 (Web Text))

Max Activating Token Index: 882

Text #6

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0512. Min Act: -0.0002

Data Index: 789805 (C4 (Web Text))

Max Activating Token Index: 46

Text #7

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0509. Min Act: -0.0002

Data Index: 903830 (C4 (Web Text))

Max Activating Token Index: 346

Text #8

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0444. Min Act: -0.0002

Data Index: 585545 (C4 (Web Text))

Max Activating Token Index: 849

Text #9

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0457. Min Act: -0.0002

Data Index: 226588 (C4 (Web Text))

Max Activating Token Index: 15

Text #10

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0492. Min Act: -0.0002

Data Index: 818461 (C4 (Web Text))

Max Activating Token Index: 442

Text #11

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0461. Min Act: -0.0002

Data Index: 345249 (C4 (Web Text))

Max Activating Token Index: 14

Text #12

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0494. Min Act: -0.0002

Data Index: 916472 (C4 (Web Text))

Max Activating Token Index: 4

Text #13

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0478. Min Act: -0.0002

Data Index: 107112 (C4 (Web Text))

Max Activating Token Index: 17

Text #14

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0458. Min Act: -0.0002

Data Index: 512495 (C4 (Web Text))

Max Activating Token Index: 1005

Text #15

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0460. Min Act: -0.0002

Data Index: 1159211 (C4 (Web Text))

Max Activating Token Index: 265

Text #16

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0355. Min Act: -0.0002

Data Index: 62119 (C4 (Web Text))

Max Activating Token Index: 163

Text #17

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0421. Min Act: -0.0002

Data Index: 1101382 (C4 (Web Text))

Max Activating Token Index: 277

Text #18

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0401. Min Act: -0.0002

Data Index: 214730 (C4 (Web Text))

Max Activating Token Index: 767

Text #19

Max Range: 0.0660. Min Range: -0.0660

Max Act: 0.0446. Min Act: -0.0002

Data Index: 529901 (C4 (Web Text))

Max Activating Token Index: 428