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

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

Neuron 1990 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0087. Min Act: -0.0003

Data Index: 275365 (C4 (Web Text))

Max Activating Token Index: 754

Text #1

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0063. Min Act: -0.0003

Data Index: 1059795 (C4 (Web Text))

Max Activating Token Index: 740

Text #2

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0062. Min Act: -0.0003

Data Index: 1400857 (Python Code)

Max Activating Token Index: 42

Text #3

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0059. Min Act: -0.0003

Data Index: 1295083 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0049. Min Act: -0.0003

Data Index: 573440 (C4 (Web Text))

Max Activating Token Index: 330

Text #5

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0053. Min Act: -0.0003

Data Index: 1580279 (Python Code)

Max Activating Token Index: 15

Text #6

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0046. Min Act: -0.0003

Data Index: 1543830 (Python Code)

Max Activating Token Index: 711

Text #7

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0050. Min Act: -0.0003

Data Index: 231021 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0040. Min Act: -0.0003

Data Index: 1471366 (Python Code)

Max Activating Token Index: 714

Text #9

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0043. Min Act: -0.0003

Data Index: 1422602 (Python Code)

Max Activating Token Index: 764

Text #10

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0041. Min Act: -0.0003

Data Index: 46306 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0040. Min Act: -0.0003

Data Index: 426457 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0039. Min Act: -0.0003

Data Index: 502153 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0038. Min Act: -0.0003

Data Index: 888144 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0037. Min Act: -0.0003

Data Index: 807912 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0035. Min Act: -0.0003

Data Index: 737494 (C4 (Web Text))

Max Activating Token Index: 1023

Text #16

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0037. Min Act: -0.0003

Data Index: 992510 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0038. Min Act: -0.0003

Data Index: 446473 (C4 (Web Text))

Max Activating Token Index: 682

Text #18

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0038. Min Act: -0.0003

Data Index: 1287914 (C4 (Web Text))

Max Activating Token Index: 1023

Text #19

Max Range: 0.0087. Min Range: -0.0087

Max Act: 0.0039. Min Act: -0.0003

Data Index: 1407137 (Python Code)

Max Activating Token Index: 1023