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

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

Neuron 3639 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.2160. Min Act: -0.0001

Data Index: 359343 (C4 (Web Text))

Max Activating Token Index: 19

Text #1

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1875. Min Act: -0.0001

Data Index: 471955 (C4 (Web Text))

Max Activating Token Index: 961

Text #2

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1994. Min Act: -0.0001

Data Index: 832420 (C4 (Web Text))

Max Activating Token Index: 657

Text #3

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1895. Min Act: -0.0001

Data Index: 382688 (C4 (Web Text))

Max Activating Token Index: 41

Text #4

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1930. Min Act: -0.0001

Data Index: 759959 (C4 (Web Text))

Max Activating Token Index: 113

Text #5

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1759. Min Act: -0.0000

Data Index: 1645818 (Python Code)

Max Activating Token Index: 161

Text #6

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1792. Min Act: -0.0001

Data Index: 849408 (C4 (Web Text))

Max Activating Token Index: 126

Text #7

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1756. Min Act: -0.0001

Data Index: 1175221 (C4 (Web Text))

Max Activating Token Index: 749

Text #8

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1740. Min Act: -0.0000

Data Index: 1653323 (Python Code)

Max Activating Token Index: 265

Text #9

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1669. Min Act: -0.0001

Data Index: 594648 (C4 (Web Text))

Max Activating Token Index: 988

Text #10

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1650. Min Act: -0.0000

Data Index: 1578312 (Python Code)

Max Activating Token Index: 137

Text #11

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1734. Min Act: -0.0001

Data Index: 673585 (C4 (Web Text))

Max Activating Token Index: 634

Text #12

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1703. Min Act: -0.0001

Data Index: 1057214 (C4 (Web Text))

Max Activating Token Index: 70

Text #13

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1727. Min Act: -0.0000

Data Index: 1459973 (Python Code)

Max Activating Token Index: 91

Text #14

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1582. Min Act: -0.0001

Data Index: 678087 (C4 (Web Text))

Max Activating Token Index: 848

Text #15

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1527. Min Act: -0.0001

Data Index: 1067596 (C4 (Web Text))

Max Activating Token Index: 187

Text #16

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1658. Min Act: -0.0000

Data Index: 1486401 (Python Code)

Max Activating Token Index: 161

Text #17

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1602. Min Act: -0.0000

Data Index: 1555970 (Python Code)

Max Activating Token Index: 196

Text #18

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1566. Min Act: -0.0000

Data Index: 1619093 (Python Code)

Max Activating Token Index: 114

Text #19

Max Range: 0.2160. Min Range: -0.2160

Max Act: 0.1605. Min Act: -0.0001

Data Index: 32702 (C4 (Web Text))

Max Activating Token Index: 163