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

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

Neuron 214 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.1153. Min Act: -0.0002

Data Index: 128129 (C4 (Web Text))

Max Activating Token Index: 798

Text #1

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.1010. Min Act: -0.0001

Data Index: 883473 (C4 (Web Text))

Max Activating Token Index: 346

Text #2

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.1033. Min Act: -0.0002

Data Index: 312206 (C4 (Web Text))

Max Activating Token Index: 306

Text #3

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.1021. Min Act: -0.0002

Data Index: 595735 (C4 (Web Text))

Max Activating Token Index: 327

Text #4

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0884. Min Act: -0.0002

Data Index: 1263760 (C4 (Web Text))

Max Activating Token Index: 50

Text #5

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0864. Min Act: -0.0001

Data Index: 1237001 (C4 (Web Text))

Max Activating Token Index: 907

Text #6

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0911. Min Act: -0.0001

Data Index: 31289 (C4 (Web Text))

Max Activating Token Index: 307

Text #7

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0765. Min Act: -0.0001

Data Index: 1545632 (Python Code)

Max Activating Token Index: 614

Text #8

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0901. Min Act: -0.0002

Data Index: 727260 (C4 (Web Text))

Max Activating Token Index: 458

Text #9

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0894. Min Act: -0.0001

Data Index: 1590801 (Python Code)

Max Activating Token Index: 941

Text #10

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0880. Min Act: -0.0002

Data Index: 387560 (C4 (Web Text))

Max Activating Token Index: 822

Text #11

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0840. Min Act: -0.0001

Data Index: 1474918 (Python Code)

Max Activating Token Index: 697

Text #12

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0840. Min Act: -0.0001

Data Index: 780495 (C4 (Web Text))

Max Activating Token Index: 228

Text #13

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0775. Min Act: -0.0001

Data Index: 1414677 (Python Code)

Max Activating Token Index: 797

Text #14

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0868. Min Act: -0.0002

Data Index: 1317283 (C4 (Web Text))

Max Activating Token Index: 964

Text #15

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0838. Min Act: -0.0001

Data Index: 1160537 (C4 (Web Text))

Max Activating Token Index: 603

Text #16

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0816. Min Act: -0.0002

Data Index: 742053 (C4 (Web Text))

Max Activating Token Index: 8

Text #17

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0811. Min Act: -0.0001

Data Index: 945707 (C4 (Web Text))

Max Activating Token Index: 401

Text #18

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0789. Min Act: -0.0001

Data Index: 689968 (C4 (Web Text))

Max Activating Token Index: 868

Text #19

Max Range: 0.1153. Min Range: -0.1153

Max Act: 0.0818. Min Act: -0.0001

Data Index: 595270 (C4 (Web Text))

Max Activating Token Index: 412