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

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

Neuron 3673 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1115. Min Range: -3.1115

Max Act: 3.1115. Min Act: -0.0000

Data Index: 562012 (C4 (Web Text))

Max Activating Token Index: 613

Text #1

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.2739. Min Act: -0.0000

Data Index: 110092 (C4 (Web Text))

Max Activating Token Index: 261

Text #2

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.1955. Min Act: -0.0000

Data Index: 586427 (C4 (Web Text))

Max Activating Token Index: 295

Text #3

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.2012. Min Act: -0.0000

Data Index: 963647 (C4 (Web Text))

Max Activating Token Index: 559

Text #4

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.0453. Min Act: -0.0000

Data Index: 307887 (C4 (Web Text))

Max Activating Token Index: 658

Text #5

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.1307. Min Act: -0.0000

Data Index: 1525096 (Python Code)

Max Activating Token Index: 689

Text #6

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.1317. Min Act: -0.0000

Data Index: 1355072 (C4 (Web Text))

Max Activating Token Index: 647

Text #7

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.1087. Min Act: -0.0000

Data Index: 167080 (C4 (Web Text))

Max Activating Token Index: 385

Text #8

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.0283. Min Act: -0.0000

Data Index: 248691 (C4 (Web Text))

Max Activating Token Index: 1019

Text #9

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.0013. Min Act: -0.0000

Data Index: 946117 (C4 (Web Text))

Max Activating Token Index: 646

Text #10

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.1126. Min Act: -0.0000

Data Index: 1333590 (C4 (Web Text))

Max Activating Token Index: 760

Text #11

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.8770. Min Act: -0.0000

Data Index: 1050994 (C4 (Web Text))

Max Activating Token Index: 136

Text #12

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.9614. Min Act: -0.0000

Data Index: 1219000 (C4 (Web Text))

Max Activating Token Index: 797

Text #13

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.8184. Min Act: -0.0000

Data Index: 1468474 (Python Code)

Max Activating Token Index: 4

Text #14

Max Range: 3.1115. Min Range: -3.1115

Max Act: 2.0039. Min Act: -0.0000

Data Index: 495615 (C4 (Web Text))

Max Activating Token Index: 93

Text #15

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.7562. Min Act: -0.0000

Data Index: 755801 (C4 (Web Text))

Max Activating Token Index: 831

Text #16

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.8637. Min Act: -0.0000

Data Index: 796945 (C4 (Web Text))

Max Activating Token Index: 822

Text #17

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.7499. Min Act: -0.0000

Data Index: 220241 (C4 (Web Text))

Max Activating Token Index: 908

Text #18

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.8179. Min Act: -0.0000

Data Index: 51969 (C4 (Web Text))

Max Activating Token Index: 883

Text #19

Max Range: 3.1115. Min Range: -3.1115

Max Act: 1.8169. Min Act: -0.0000

Data Index: 754403 (C4 (Web Text))

Max Activating Token Index: 257