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

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

Neuron 3596 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.5233. Min Act: -0.0001

Data Index: 881561 (C4 (Web Text))

Max Activating Token Index: 867

Text #1

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.4951. Min Act: -0.0001

Data Index: 1002700 (C4 (Web Text))

Max Activating Token Index: 695

Text #2

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.2900. Min Act: -0.0001

Data Index: 986867 (C4 (Web Text))

Max Activating Token Index: 388

Text #3

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.2109. Min Act: -0.0001

Data Index: 1472223 (Python Code)

Max Activating Token Index: 34

Text #4

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1882. Min Act: -0.0001

Data Index: 366714 (C4 (Web Text))

Max Activating Token Index: 405

Text #5

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1427. Min Act: -0.0001

Data Index: 1492806 (Python Code)

Max Activating Token Index: 10

Text #6

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.2359. Min Act: -0.0001

Data Index: 472613 (C4 (Web Text))

Max Activating Token Index: 350

Text #7

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1565. Min Act: -0.0001

Data Index: 714595 (C4 (Web Text))

Max Activating Token Index: 96

Text #8

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1030. Min Act: -0.0001

Data Index: 1198685 (C4 (Web Text))

Max Activating Token Index: 26

Text #9

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1688. Min Act: -0.0001

Data Index: 1088655 (C4 (Web Text))

Max Activating Token Index: 982

Text #10

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1686. Min Act: -0.0001

Data Index: 1458351 (Python Code)

Max Activating Token Index: 606

Text #11

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.0211. Min Act: -0.0001

Data Index: 113829 (C4 (Web Text))

Max Activating Token Index: 737

Text #12

Max Range: 1.5233. Min Range: -1.5233

Max Act: 0.9821. Min Act: -0.0001

Data Index: 880362 (C4 (Web Text))

Max Activating Token Index: 867

Text #13

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.0984. Min Act: -0.0001

Data Index: 1506168 (Python Code)

Max Activating Token Index: 430

Text #14

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.0949. Min Act: -0.0001

Data Index: 424183 (C4 (Web Text))

Max Activating Token Index: 270

Text #15

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1284. Min Act: -0.0001

Data Index: 1168524 (C4 (Web Text))

Max Activating Token Index: 168

Text #16

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1140. Min Act: -0.0001

Data Index: 640821 (C4 (Web Text))

Max Activating Token Index: 957

Text #17

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.1283. Min Act: -0.0001

Data Index: 1034507 (C4 (Web Text))

Max Activating Token Index: 165

Text #18

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.0759. Min Act: -0.0001

Data Index: 393055 (C4 (Web Text))

Max Activating Token Index: 389

Text #19

Max Range: 1.5233. Min Range: -1.5233

Max Act: 1.0108. Min Act: -0.0001

Data Index: 1426007 (Python Code)

Max Activating Token Index: 215