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

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

Neuron 1951 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: 0.0009. Min Range: -0.0009

Max Act: 0.0009. Min Act: -0.0006

Data Index: 1011833 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 1570985 (Python Code)

Max Activating Token Index: 524

Text #2

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 1293237 (C4 (Web Text))

Max Activating Token Index: 1023

Text #3

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 984882 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 1168407 (C4 (Web Text))

Max Activating Token Index: 1023

Text #5

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0007

Data Index: 1181548 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 58866 (C4 (Web Text))

Max Activating Token Index: 495

Text #7

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 997039 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0007. Min Act: -0.0006

Data Index: 375607 (C4 (Web Text))

Max Activating Token Index: 590

Text #9

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 1018078 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0007

Data Index: 79273 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 1094592 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 212265 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 1408547 (Python Code)

Max Activating Token Index: 363

Text #14

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 226795 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 202364 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 932800 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 129611 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 289165 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.0009. Min Range: -0.0009

Max Act: 0.0006. Min Act: -0.0006

Data Index: 282979 (C4 (Web Text))

Max Activating Token Index: 379