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

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

Neuron 627 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0150. Min Act: -0.0005

Data Index: 436093 (C4 (Web Text))

Max Activating Token Index: 394

Text #1

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0128. Min Act: -0.0006

Data Index: 109589 (C4 (Web Text))

Max Activating Token Index: 344

Text #2

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0128. Min Act: -0.0005

Data Index: 501030 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0131. Min Act: -0.0006

Data Index: 1580820 (Python Code)

Max Activating Token Index: 1023

Text #4

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0133. Min Act: -0.0005

Data Index: 831206 (C4 (Web Text))

Max Activating Token Index: 583

Text #5

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0122. Min Act: -0.0006

Data Index: 923221 (C4 (Web Text))

Max Activating Token Index: 934

Text #6

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0127. Min Act: -0.0006

Data Index: 127054 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0131. Min Act: -0.0005

Data Index: 360563 (C4 (Web Text))

Max Activating Token Index: 797

Text #8

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0125. Min Act: -0.0006

Data Index: 644545 (C4 (Web Text))

Max Activating Token Index: 466

Text #9

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0128. Min Act: -0.0005

Data Index: 433488 (C4 (Web Text))

Max Activating Token Index: 609

Text #10

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0133. Min Act: -0.0006

Data Index: 1503398 (Python Code)

Max Activating Token Index: 1023

Text #11

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0119. Min Act: -0.0006

Data Index: 224769 (C4 (Web Text))

Max Activating Token Index: 853

Text #12

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0123. Min Act: -0.0006

Data Index: 1402053 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0115. Min Act: -0.0006

Data Index: 106745 (C4 (Web Text))

Max Activating Token Index: 502

Text #14

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0119. Min Act: -0.0005

Data Index: 1306877 (C4 (Web Text))

Max Activating Token Index: 360

Text #15

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0110. Min Act: -0.0006

Data Index: 1226003 (C4 (Web Text))

Max Activating Token Index: 315

Text #16

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0118. Min Act: -0.0005

Data Index: 1319524 (C4 (Web Text))

Max Activating Token Index: 599

Text #17

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0116. Min Act: -0.0005

Data Index: 908014 (C4 (Web Text))

Max Activating Token Index: 387

Text #18

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0121. Min Act: -0.0005

Data Index: 186877 (C4 (Web Text))

Max Activating Token Index: 898

Text #19

Max Range: 0.0150. Min Range: -0.0150

Max Act: 0.0113. Min Act: -0.0005

Data Index: 230833 (C4 (Web Text))

Max Activating Token Index: 168