Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5563 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.5148. Min Act: -0.0000

Data Index: 874661 (C4 (Web Text))

Max Activating Token Index: 117

Text #1

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.5172. Min Act: -0.0000

Data Index: 557723 (C4 (Web Text))

Max Activating Token Index: 940

Text #2

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.4146. Min Act: -0.0000

Data Index: 840863 (C4 (Web Text))

Max Activating Token Index: 130

Text #3

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.5414. Min Act: -0.0000

Data Index: 800315 (C4 (Web Text))

Max Activating Token Index: 122

Text #4

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.5142. Min Act: -0.0000

Data Index: 519504 (C4 (Web Text))

Max Activating Token Index: 865

Text #5

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.3745. Min Act: -0.0000

Data Index: 268312 (C4 (Web Text))

Max Activating Token Index: 943

Text #6

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.3506. Min Act: -0.0000

Data Index: 1424818 (Python Code)

Max Activating Token Index: 82

Text #7

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.2920. Min Act: -0.0000

Data Index: 912178 (C4 (Web Text))

Max Activating Token Index: 350

Text #8

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1610. Min Act: -0.0000

Data Index: 1227540 (C4 (Web Text))

Max Activating Token Index: 950

Text #9

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.2663. Min Act: -0.0000

Data Index: 389051 (C4 (Web Text))

Max Activating Token Index: 1013

Text #10

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1358. Min Act: -0.0000

Data Index: 433748 (C4 (Web Text))

Max Activating Token Index: 493

Text #11

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1754. Min Act: -0.0000

Data Index: 861200 (C4 (Web Text))

Max Activating Token Index: 755

Text #12

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1818. Min Act: -0.0000

Data Index: 814325 (C4 (Web Text))

Max Activating Token Index: 725

Text #13

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1664. Min Act: -0.0000

Data Index: 1550325 (Python Code)

Max Activating Token Index: 1010

Text #14

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1243. Min Act: -0.0000

Data Index: 108767 (C4 (Web Text))

Max Activating Token Index: 405

Text #15

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1651. Min Act: -0.0000

Data Index: 1227141 (C4 (Web Text))

Max Activating Token Index: 1012

Text #16

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.0135. Min Act: -0.0000

Data Index: 1243927 (C4 (Web Text))

Max Activating Token Index: 546

Text #17

Max Range: 2.5414. Min Range: -2.5414

Max Act: 2.1249. Min Act: -0.0000

Data Index: 439923 (C4 (Web Text))

Max Activating Token Index: 447

Text #18

Max Range: 2.5414. Min Range: -2.5414

Max Act: 1.9089. Min Act: -0.0000

Data Index: 815330 (C4 (Web Text))

Max Activating Token Index: 821

Text #19

Max Range: 2.5414. Min Range: -2.5414

Max Act: 1.7924. Min Act: -0.0000

Data Index: 307046 (C4 (Web Text))

Max Activating Token Index: 886