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

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

Neuron 1615 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.5388. Min Act: -0.0000

Data Index: 580218 (C4 (Web Text))

Max Activating Token Index: 22

Text #1

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.5677. Min Act: -0.0000

Data Index: 478424 (C4 (Web Text))

Max Activating Token Index: 40

Text #2

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.3887. Min Act: -0.0000

Data Index: 215403 (C4 (Web Text))

Max Activating Token Index: 96

Text #3

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.5287. Min Act: -0.0001

Data Index: 213689 (C4 (Web Text))

Max Activating Token Index: 353

Text #4

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.4117. Min Act: -0.0001

Data Index: 249711 (C4 (Web Text))

Max Activating Token Index: 442

Text #5

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.4292. Min Act: -0.0001

Data Index: 1500150 (Python Code)

Max Activating Token Index: 952

Text #6

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2041. Min Act: -0.0000

Data Index: 406360 (C4 (Web Text))

Max Activating Token Index: 314

Text #7

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2907. Min Act: -0.0000

Data Index: 401903 (C4 (Web Text))

Max Activating Token Index: 691

Text #8

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.4627. Min Act: -0.0000

Data Index: 730006 (C4 (Web Text))

Max Activating Token Index: 361

Text #9

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.3702. Min Act: -0.0001

Data Index: 1034009 (C4 (Web Text))

Max Activating Token Index: 3

Text #10

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.4073. Min Act: -0.0001

Data Index: 990515 (C4 (Web Text))

Max Activating Token Index: 658

Text #11

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.3211. Min Act: -0.0001

Data Index: 259717 (C4 (Web Text))

Max Activating Token Index: 652

Text #12

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.3411. Min Act: -0.0000

Data Index: 1492569 (Python Code)

Max Activating Token Index: 964

Text #13

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2872. Min Act: -0.0000

Data Index: 680354 (C4 (Web Text))

Max Activating Token Index: 59

Text #14

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2408. Min Act: -0.0001

Data Index: 790334 (C4 (Web Text))

Max Activating Token Index: 121

Text #15

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2756. Min Act: -0.0000

Data Index: 709014 (C4 (Web Text))

Max Activating Token Index: 5

Text #16

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2941. Min Act: -0.0000

Data Index: 1656928 (Python Code)

Max Activating Token Index: 524

Text #17

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2378. Min Act: -0.0000

Data Index: 1085204 (C4 (Web Text))

Max Activating Token Index: 585

Text #18

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.4594. Min Act: -0.0001

Data Index: 758764 (C4 (Web Text))

Max Activating Token Index: 283

Text #19

Max Range: 1.5677. Min Range: -1.5677

Max Act: 1.2939. Min Act: -0.0000

Data Index: 467443 (C4 (Web Text))

Max Activating Token Index: 62