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

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

Neuron 1059 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.9755. Min Range: -5.9755

Max Act: 5.9755. Min Act: -0.0000

Data Index: 865703 (C4 (Web Text))

Max Activating Token Index: 667

Text #1

Max Range: 5.9755. Min Range: -5.9755

Max Act: 5.5058. Min Act: -0.0000

Data Index: 20708 (C4 (Web Text))

Max Activating Token Index: 139

Text #2

Max Range: 5.9755. Min Range: -5.9755

Max Act: 5.1610. Min Act: -0.0000

Data Index: 20644 (C4 (Web Text))

Max Activating Token Index: 259

Text #3

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.8019. Min Act: -0.0000

Data Index: 635168 (C4 (Web Text))

Max Activating Token Index: 800

Text #4

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.6310. Min Act: -0.0000

Data Index: 231660 (C4 (Web Text))

Max Activating Token Index: 166

Text #5

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.7897. Min Act: -0.0000

Data Index: 886508 (C4 (Web Text))

Max Activating Token Index: 127

Text #6

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.4785. Min Act: -0.0000

Data Index: 1307297 (C4 (Web Text))

Max Activating Token Index: 331

Text #7

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.4149. Min Act: -0.0000

Data Index: 116683 (C4 (Web Text))

Max Activating Token Index: 812

Text #8

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.1471. Min Act: -0.0000

Data Index: 713477 (C4 (Web Text))

Max Activating Token Index: 192

Text #9

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.2973. Min Act: -0.0000

Data Index: 532874 (C4 (Web Text))

Max Activating Token Index: 826

Text #10

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.4555. Min Act: -0.0000

Data Index: 1035128 (C4 (Web Text))

Max Activating Token Index: 169

Text #11

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.2399. Min Act: -0.0000

Data Index: 542140 (C4 (Web Text))

Max Activating Token Index: 62

Text #12

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.2346. Min Act: -0.0000

Data Index: 120788 (C4 (Web Text))

Max Activating Token Index: 899

Text #13

Max Range: 5.9755. Min Range: -5.9755

Max Act: 3.7290. Min Act: -0.0000

Data Index: 1450300 (Python Code)

Max Activating Token Index: 606

Text #14

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.1831. Min Act: -0.0000

Data Index: 817845 (C4 (Web Text))

Max Activating Token Index: 574

Text #15

Max Range: 5.9755. Min Range: -5.9755

Max Act: 3.9971. Min Act: -0.0000

Data Index: 257478 (C4 (Web Text))

Max Activating Token Index: 450

Text #16

Max Range: 5.9755. Min Range: -5.9755

Max Act: 3.9527. Min Act: -0.0000

Data Index: 851157 (C4 (Web Text))

Max Activating Token Index: 78

Text #17

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.1509. Min Act: -0.0000

Data Index: 412681 (C4 (Web Text))

Max Activating Token Index: 406

Text #18

Max Range: 5.9755. Min Range: -5.9755

Max Act: 3.7986. Min Act: -0.0000

Data Index: 1128206 (C4 (Web Text))

Max Activating Token Index: 751

Text #19

Max Range: 5.9755. Min Range: -5.9755

Max Act: 4.2956. Min Act: -0.0000

Data Index: 713664 (C4 (Web Text))

Max Activating Token Index: 911