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

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

Neuron 642 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: 0.1070. Min Range: -0.1070

Max Act: 0.1070. Min Act: -0.0002

Data Index: 440829 (C4 (Web Text))

Max Activating Token Index: 248

Text #1

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0950. Min Act: -0.0002

Data Index: 644136 (C4 (Web Text))

Max Activating Token Index: 71

Text #2

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0945. Min Act: -0.0001

Data Index: 968238 (C4 (Web Text))

Max Activating Token Index: 426

Text #3

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.1007. Min Act: -0.0001

Data Index: 33790 (C4 (Web Text))

Max Activating Token Index: 403

Text #4

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0902. Min Act: -0.0002

Data Index: 666530 (C4 (Web Text))

Max Activating Token Index: 118

Text #5

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0882. Min Act: -0.0001

Data Index: 986319 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0910. Min Act: -0.0002

Data Index: 28579 (C4 (Web Text))

Max Activating Token Index: 1006

Text #7

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0960. Min Act: -0.0001

Data Index: 545609 (C4 (Web Text))

Max Activating Token Index: 722

Text #8

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0768. Min Act: -0.0001

Data Index: 174299 (C4 (Web Text))

Max Activating Token Index: 835

Text #9

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0734. Min Act: -0.0001

Data Index: 802501 (C4 (Web Text))

Max Activating Token Index: 821

Text #10

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0804. Min Act: -0.0001

Data Index: 108297 (C4 (Web Text))

Max Activating Token Index: 947

Text #11

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0722. Min Act: -0.0001

Data Index: 1624312 (Python Code)

Max Activating Token Index: 252

Text #12

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0675. Min Act: -0.0001

Data Index: 104908 (C4 (Web Text))

Max Activating Token Index: 534

Text #13

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0721. Min Act: -0.0002

Data Index: 622733 (C4 (Web Text))

Max Activating Token Index: 338

Text #14

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0729. Min Act: -0.0002

Data Index: 1346909 (C4 (Web Text))

Max Activating Token Index: 973

Text #15

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0657. Min Act: -0.0002

Data Index: 1079172 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0724. Min Act: -0.0001

Data Index: 1184166 (C4 (Web Text))

Max Activating Token Index: 888

Text #17

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0641. Min Act: -0.0002

Data Index: 368131 (C4 (Web Text))

Max Activating Token Index: 539

Text #18

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0682. Min Act: -0.0002

Data Index: 420769 (C4 (Web Text))

Max Activating Token Index: 893

Text #19

Max Range: 0.1070. Min Range: -0.1070

Max Act: 0.0655. Min Act: -0.0001

Data Index: 469430 (C4 (Web Text))

Max Activating Token Index: 744