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

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

Neuron 5866 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: 0.6669. Min Range: -0.6669

Max Act: 0.6669. Min Act: -0.0000

Data Index: 349023 (C4 (Web Text))

Max Activating Token Index: 913

Text #1

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.6565. Min Act: -0.0000

Data Index: 1034706 (C4 (Web Text))

Max Activating Token Index: 17

Text #2

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.6338. Min Act: -0.0000

Data Index: 454470 (C4 (Web Text))

Max Activating Token Index: 726

Text #3

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.6342. Min Act: -0.0000

Data Index: 308205 (C4 (Web Text))

Max Activating Token Index: 927

Text #4

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.5607. Min Act: -0.0000

Data Index: 167552 (C4 (Web Text))

Max Activating Token Index: 158

Text #5

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.5756. Min Act: -0.0000

Data Index: 624560 (C4 (Web Text))

Max Activating Token Index: 36

Text #6

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.5304. Min Act: -0.0000

Data Index: 914399 (C4 (Web Text))

Max Activating Token Index: 771

Text #7

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.5719. Min Act: -0.0000

Data Index: 1514800 (Python Code)

Max Activating Token Index: 244

Text #8

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4859. Min Act: -0.0000

Data Index: 697961 (C4 (Web Text))

Max Activating Token Index: 844

Text #9

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4973. Min Act: -0.0000

Data Index: 1545294 (Python Code)

Max Activating Token Index: 975

Text #10

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4977. Min Act: -0.0000

Data Index: 546063 (C4 (Web Text))

Max Activating Token Index: 596

Text #11

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4918. Min Act: -0.0000

Data Index: 1563334 (Python Code)

Max Activating Token Index: 487

Text #12

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4936. Min Act: -0.0000

Data Index: 540521 (C4 (Web Text))

Max Activating Token Index: 721

Text #13

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.5023. Min Act: -0.0000

Data Index: 810973 (C4 (Web Text))

Max Activating Token Index: 289

Text #14

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4474. Min Act: -0.0000

Data Index: 1461657 (Python Code)

Max Activating Token Index: 25

Text #15

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4526. Min Act: -0.0000

Data Index: 36951 (C4 (Web Text))

Max Activating Token Index: 1016

Text #16

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4603. Min Act: -0.0000

Data Index: 441112 (C4 (Web Text))

Max Activating Token Index: 864

Text #17

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4253. Min Act: -0.0000

Data Index: 1055533 (C4 (Web Text))

Max Activating Token Index: 592

Text #18

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.3871. Min Act: -0.0000

Data Index: 1123671 (C4 (Web Text))

Max Activating Token Index: 881

Text #19

Max Range: 0.6669. Min Range: -0.6669

Max Act: 0.4137. Min Act: -0.0000

Data Index: 1010169 (C4 (Web Text))

Max Activating Token Index: 47