Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2576 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.7586. Min Act: -0.0002

Data Index: 789861 (C4 (Web Text))

Max Activating Token Index: 936

Text #1

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.7272. Min Act: -0.0002

Data Index: 992251 (C4 (Web Text))

Max Activating Token Index: 315

Text #2

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.3969. Min Act: -0.0002

Data Index: 1006437 (C4 (Web Text))

Max Activating Token Index: 333

Text #3

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.2228. Min Act: -0.0002

Data Index: 544595 (C4 (Web Text))

Max Activating Token Index: 723

Text #4

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.3198. Min Act: -0.0002

Data Index: 1656764 (Python Code)

Max Activating Token Index: 232

Text #5

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.2040. Min Act: -0.0002

Data Index: 651533 (C4 (Web Text))

Max Activating Token Index: 924

Text #6

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.2542. Min Act: -0.0002

Data Index: 1330805 (C4 (Web Text))

Max Activating Token Index: 790

Text #7

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.2128. Min Act: -0.0002

Data Index: 1386069 (Python Code)

Max Activating Token Index: 481

Text #8

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.1588. Min Act: -0.0002

Data Index: 1489981 (Python Code)

Max Activating Token Index: 412

Text #9

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.1163. Min Act: -0.0002

Data Index: 214416 (C4 (Web Text))

Max Activating Token Index: 530

Text #10

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.1354. Min Act: -0.0002

Data Index: 823219 (C4 (Web Text))

Max Activating Token Index: 40

Text #11

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.1478. Min Act: -0.0002

Data Index: 734698 (C4 (Web Text))

Max Activating Token Index: 783

Text #12

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0773. Min Act: -0.0002

Data Index: 447057 (C4 (Web Text))

Max Activating Token Index: 83

Text #13

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.1392. Min Act: -0.0002

Data Index: 701427 (C4 (Web Text))

Max Activating Token Index: 947

Text #14

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0847. Min Act: -0.0002

Data Index: 1438836 (Python Code)

Max Activating Token Index: 800

Text #15

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0947. Min Act: -0.0002

Data Index: 1060814 (C4 (Web Text))

Max Activating Token Index: 28

Text #16

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0600. Min Act: -0.0002

Data Index: 1447087 (Python Code)

Max Activating Token Index: 24

Text #17

Max Range: 1.7586. Min Range: -1.7586

Max Act: 0.9952. Min Act: -0.0002

Data Index: 210563 (C4 (Web Text))

Max Activating Token Index: 227

Text #18

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0532. Min Act: -0.0002

Data Index: 322040 (C4 (Web Text))

Max Activating Token Index: 65

Text #19

Max Range: 1.7586. Min Range: -1.7586

Max Act: 1.0416. Min Act: -0.0002

Data Index: 2280 (C4 (Web Text))

Max Activating Token Index: 1008