Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1600 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.8533. Min Act: -0.0000

Data Index: 835943 (C4 (Web Text))

Max Activating Token Index: 628

Text #1

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.5072. Min Act: -0.0000

Data Index: 201488 (C4 (Web Text))

Max Activating Token Index: 970

Text #2

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.4044. Min Act: -0.0000

Data Index: 1516052 (Python Code)

Max Activating Token Index: 232

Text #3

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.4921. Min Act: -0.0000

Data Index: 1372547 (Python Code)

Max Activating Token Index: 226

Text #4

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.1237. Min Act: -0.0000

Data Index: 1398715 (Python Code)

Max Activating Token Index: 714

Text #5

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.2389. Min Act: -0.0000

Data Index: 627022 (C4 (Web Text))

Max Activating Token Index: 503

Text #6

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.0000. Min Act: -0.0000

Data Index: 1573943 (Python Code)

Max Activating Token Index: 791

Text #7

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.2846. Min Act: -0.0000

Data Index: 1285701 (C4 (Web Text))

Max Activating Token Index: 127

Text #8

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.3726. Min Act: -0.0000

Data Index: 1582054 (Python Code)

Max Activating Token Index: 729

Text #9

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.3620. Min Act: -0.0000

Data Index: 1632391 (Python Code)

Max Activating Token Index: 113

Text #10

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.0391. Min Act: -0.0000

Data Index: 1587055 (Python Code)

Max Activating Token Index: 153

Text #11

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.1297. Min Act: -0.0000

Data Index: 1559843 (Python Code)

Max Activating Token Index: 532

Text #12

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.2212. Min Act: -0.0000

Data Index: 377409 (C4 (Web Text))

Max Activating Token Index: 703

Text #13

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.0545. Min Act: -0.0000

Data Index: 579623 (C4 (Web Text))

Max Activating Token Index: 281

Text #14

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.1450. Min Act: -0.0000

Data Index: 1485094 (Python Code)

Max Activating Token Index: 354

Text #15

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.1048. Min Act: -0.0000

Data Index: 1585986 (Python Code)

Max Activating Token Index: 552

Text #16

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.2269. Min Act: -0.0000

Data Index: 1647579 (Python Code)

Max Activating Token Index: 864

Text #17

Max Range: 2.8533. Min Range: -2.8533

Max Act: 1.9832. Min Act: -0.0000

Data Index: 1286358 (C4 (Web Text))

Max Activating Token Index: 97

Text #18

Max Range: 2.8533. Min Range: -2.8533

Max Act: 2.0575. Min Act: -0.0000

Data Index: 35089 (C4 (Web Text))

Max Activating Token Index: 556

Text #19

Max Range: 2.8533. Min Range: -2.8533

Max Act: 1.9525. Min Act: -0.0000

Data Index: 1073621 (C4 (Web Text))

Max Activating Token Index: 251