Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1465 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1351. Min Act: -0.0005

Data Index: 433156 (C4 (Web Text))

Max Activating Token Index: 430

Text #1

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1411. Min Act: -0.0005

Data Index: 767398 (C4 (Web Text))

Max Activating Token Index: 1006

Text #2

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1303. Min Act: -0.0006

Data Index: 954980 (C4 (Web Text))

Max Activating Token Index: 325

Text #3

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1182. Min Act: -0.0005

Data Index: 705546 (C4 (Web Text))

Max Activating Token Index: 289

Text #4

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1175. Min Act: -0.0005

Data Index: 582570 (C4 (Web Text))

Max Activating Token Index: 697

Text #5

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1174. Min Act: -0.0005

Data Index: 641726 (C4 (Web Text))

Max Activating Token Index: 374

Text #6

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1168. Min Act: -0.0006

Data Index: 1126696 (C4 (Web Text))

Max Activating Token Index: 856

Text #7

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1124. Min Act: -0.0005

Data Index: 1210054 (C4 (Web Text))

Max Activating Token Index: 22

Text #8

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1068. Min Act: -0.0005

Data Index: 1586841 (Python Code)

Max Activating Token Index: 626

Text #9

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1033. Min Act: -0.0005

Data Index: 1075167 (C4 (Web Text))

Max Activating Token Index: 277

Text #10

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1081. Min Act: -0.0005

Data Index: 390439 (C4 (Web Text))

Max Activating Token Index: 726

Text #11

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0991. Min Act: -0.0005

Data Index: 1405953 (Python Code)

Max Activating Token Index: 388

Text #12

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0987. Min Act: -0.0005

Data Index: 1572043 (Python Code)

Max Activating Token Index: 387

Text #13

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.1006. Min Act: -0.0005

Data Index: 907666 (C4 (Web Text))

Max Activating Token Index: 143

Text #14

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0984. Min Act: -0.0005

Data Index: 1379317 (Python Code)

Max Activating Token Index: 622

Text #15

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0981. Min Act: -0.0005

Data Index: 20036 (C4 (Web Text))

Max Activating Token Index: 184

Text #16

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0957. Min Act: -0.0005

Data Index: 822511 (C4 (Web Text))

Max Activating Token Index: 833

Text #17

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0922. Min Act: -0.0005

Data Index: 249589 (C4 (Web Text))

Max Activating Token Index: 556

Text #18

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0929. Min Act: -0.0005

Data Index: 1337419 (C4 (Web Text))

Max Activating Token Index: 623

Text #19

Max Range: 0.1411. Min Range: -0.1411

Max Act: 0.0914. Min Act: -0.0005

Data Index: 113924 (C4 (Web Text))

Max Activating Token Index: 868