Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1878 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0040. Min Act: -0.0005

Data Index: 1588229 (Python Code)

Max Activating Token Index: 668

Text #1

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0039. Min Act: -0.0006

Data Index: 679591 (C4 (Web Text))

Max Activating Token Index: 996

Text #2

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0038. Min Act: -0.0006

Data Index: 108737 (C4 (Web Text))

Max Activating Token Index: 996

Text #3

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0006

Data Index: 501507 (C4 (Web Text))

Max Activating Token Index: 653

Text #4

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0006

Data Index: 1355625 (C4 (Web Text))

Max Activating Token Index: 790

Text #5

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0006

Data Index: 994827 (C4 (Web Text))

Max Activating Token Index: 926

Text #6

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0005

Data Index: 1601130 (Python Code)

Max Activating Token Index: 767

Text #7

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0006

Data Index: 643619 (C4 (Web Text))

Max Activating Token Index: 511

Text #8

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0006

Data Index: 1568160 (Python Code)

Max Activating Token Index: 543

Text #9

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0006

Data Index: 1185459 (C4 (Web Text))

Max Activating Token Index: 988

Text #10

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0037. Min Act: -0.0005

Data Index: 904869 (C4 (Web Text))

Max Activating Token Index: 312

Text #11

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0005

Data Index: 1444569 (Python Code)

Max Activating Token Index: 589

Text #12

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0006

Data Index: 1537553 (Python Code)

Max Activating Token Index: 1019

Text #13

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0006

Data Index: 1011 (C4 (Web Text))

Max Activating Token Index: 306

Text #14

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0005

Data Index: 1542673 (Python Code)

Max Activating Token Index: 665

Text #15

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0005

Data Index: 1517455 (Python Code)

Max Activating Token Index: 767

Text #16

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0005

Data Index: 932901 (C4 (Web Text))

Max Activating Token Index: 149

Text #17

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0006

Data Index: 1464437 (Python Code)

Max Activating Token Index: 133

Text #18

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0036. Min Act: -0.0005

Data Index: 1520565 (Python Code)

Max Activating Token Index: 517

Text #19

Max Range: 0.0040. Min Range: -0.0040

Max Act: 0.0035. Min Act: -0.0005

Data Index: 1488852 (Python Code)

Max Activating Token Index: 807