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

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

Neuron 3966 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.5195. Min Act: -0.0001

Data Index: 1516210 (Python Code)

Max Activating Token Index: 648

Text #1

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.5682. Min Act: -0.0001

Data Index: 1413121 (Python Code)

Max Activating Token Index: 999

Text #2

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.5745. Min Act: -0.0002

Data Index: 733782 (C4 (Web Text))

Max Activating Token Index: 310

Text #3

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.5347. Min Act: -0.0001

Data Index: 1534946 (Python Code)

Max Activating Token Index: 732

Text #4

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4937. Min Act: -0.0002

Data Index: 1054801 (C4 (Web Text))

Max Activating Token Index: 165

Text #5

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.5209. Min Act: -0.0001

Data Index: 557208 (C4 (Web Text))

Max Activating Token Index: 221

Text #6

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4802. Min Act: -0.0001

Data Index: 1518805 (Python Code)

Max Activating Token Index: 926

Text #7

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4920. Min Act: -0.0002

Data Index: 1278865 (C4 (Web Text))

Max Activating Token Index: 253

Text #8

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4843. Min Act: -0.0001

Data Index: 1650165 (Python Code)

Max Activating Token Index: 872

Text #9

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4931. Min Act: -0.0002

Data Index: 164231 (C4 (Web Text))

Max Activating Token Index: 14

Text #10

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4547. Min Act: -0.0001

Data Index: 1378908 (Python Code)

Max Activating Token Index: 951

Text #11

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4457. Min Act: -0.0001

Data Index: 1392354 (Python Code)

Max Activating Token Index: 983

Text #12

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4405. Min Act: -0.0001

Data Index: 1432635 (Python Code)

Max Activating Token Index: 360

Text #13

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.3978. Min Act: -0.0001

Data Index: 185685 (C4 (Web Text))

Max Activating Token Index: 966

Text #14

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4394. Min Act: -0.0001

Data Index: 1453507 (Python Code)

Max Activating Token Index: 882

Text #15

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4408. Min Act: -0.0001

Data Index: 899945 (C4 (Web Text))

Max Activating Token Index: 865

Text #16

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4193. Min Act: -0.0001

Data Index: 1529625 (Python Code)

Max Activating Token Index: 472

Text #17

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.3832. Min Act: -0.0001

Data Index: 1476439 (Python Code)

Max Activating Token Index: 1020

Text #18

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4545. Min Act: -0.0001

Data Index: 1385910 (Python Code)

Max Activating Token Index: 468

Text #19

Max Range: 0.5745. Min Range: -0.5745

Max Act: 0.4117. Min Act: -0.0001

Data Index: 3830 (C4 (Web Text))

Max Activating Token Index: 955