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

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

Neuron 3000 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.6980. Min Act: -0.0000

Data Index: 7257 (C4 (Web Text))

Max Activating Token Index: 77

Text #1

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.5891. Min Act: -0.0000

Data Index: 815823 (C4 (Web Text))

Max Activating Token Index: 1018

Text #2

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.5355. Min Act: -0.0000

Data Index: 344952 (C4 (Web Text))

Max Activating Token Index: 882

Text #3

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3426. Min Act: -0.0000

Data Index: 241413 (C4 (Web Text))

Max Activating Token Index: 22

Text #4

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3605. Min Act: -0.0000

Data Index: 1431481 (Python Code)

Max Activating Token Index: 5

Text #5

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1995. Min Act: -0.0000

Data Index: 200715 (C4 (Web Text))

Max Activating Token Index: 307

Text #6

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.4003. Min Act: -0.0000

Data Index: 213097 (C4 (Web Text))

Max Activating Token Index: 677

Text #7

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3748. Min Act: -0.0000

Data Index: 880667 (C4 (Web Text))

Max Activating Token Index: 280

Text #8

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3474. Min Act: -0.0000

Data Index: 227821 (C4 (Web Text))

Max Activating Token Index: 63

Text #9

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.2804. Min Act: -0.0000

Data Index: 1092878 (C4 (Web Text))

Max Activating Token Index: 249

Text #10

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.2664. Min Act: -0.0000

Data Index: 578094 (C4 (Web Text))

Max Activating Token Index: 563

Text #11

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1750. Min Act: -0.0000

Data Index: 200639 (C4 (Web Text))

Max Activating Token Index: 736

Text #12

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3112. Min Act: -0.0000

Data Index: 150650 (C4 (Web Text))

Max Activating Token Index: 291

Text #13

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.3028. Min Act: -0.0000

Data Index: 493324 (C4 (Web Text))

Max Activating Token Index: 339

Text #14

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1558. Min Act: -0.0000

Data Index: 726755 (C4 (Web Text))

Max Activating Token Index: 153

Text #15

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1985. Min Act: -0.0000

Data Index: 1027118 (C4 (Web Text))

Max Activating Token Index: 966

Text #16

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1866. Min Act: -0.0000

Data Index: 1563221 (Python Code)

Max Activating Token Index: 718

Text #17

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.1947. Min Act: -0.0000

Data Index: 304839 (C4 (Web Text))

Max Activating Token Index: 269

Text #18

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.0597. Min Act: -0.0000

Data Index: 546940 (C4 (Web Text))

Max Activating Token Index: 333

Text #19

Max Range: 1.6980. Min Range: -1.6980

Max Act: 1.0565. Min Act: -0.0000

Data Index: 406785 (C4 (Web Text))

Max Activating Token Index: 702