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

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

Neuron 2562 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.8047. Min Act: -0.0001

Data Index: 845889 (C4 (Web Text))

Max Activating Token Index: 30

Text #1

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.8558. Min Act: -0.0001

Data Index: 1002987 (C4 (Web Text))

Max Activating Token Index: 615

Text #2

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.7777. Min Act: -0.0001

Data Index: 129295 (C4 (Web Text))

Max Activating Token Index: 922

Text #3

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.7719. Min Act: -0.0002

Data Index: 428987 (C4 (Web Text))

Max Activating Token Index: 167

Text #4

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.7251. Min Act: -0.0002

Data Index: 263011 (C4 (Web Text))

Max Activating Token Index: 949

Text #5

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6615. Min Act: -0.0001

Data Index: 256252 (C4 (Web Text))

Max Activating Token Index: 839

Text #6

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6715. Min Act: -0.0001

Data Index: 1107979 (C4 (Web Text))

Max Activating Token Index: 666

Text #7

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.7435. Min Act: -0.0001

Data Index: 637507 (C4 (Web Text))

Max Activating Token Index: 803

Text #8

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6685. Min Act: -0.0002

Data Index: 1119614 (C4 (Web Text))

Max Activating Token Index: 683

Text #9

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6470. Min Act: -0.0001

Data Index: 1581613 (Python Code)

Max Activating Token Index: 480

Text #10

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6464. Min Act: -0.0001

Data Index: 825870 (C4 (Web Text))

Max Activating Token Index: 25

Text #11

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6197. Min Act: -0.0001

Data Index: 532017 (C4 (Web Text))

Max Activating Token Index: 967

Text #12

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6257. Min Act: -0.0001

Data Index: 628959 (C4 (Web Text))

Max Activating Token Index: 365

Text #13

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6074. Min Act: -0.0001

Data Index: 692300 (C4 (Web Text))

Max Activating Token Index: 467

Text #14

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6064. Min Act: -0.0001

Data Index: 896796 (C4 (Web Text))

Max Activating Token Index: 350

Text #15

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6367. Min Act: -0.0002

Data Index: 23005 (C4 (Web Text))

Max Activating Token Index: 452

Text #16

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.5947. Min Act: -0.0002

Data Index: 869139 (C4 (Web Text))

Max Activating Token Index: 915

Text #17

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6125. Min Act: -0.0001

Data Index: 1316647 (C4 (Web Text))

Max Activating Token Index: 165

Text #18

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6121. Min Act: -0.0002

Data Index: 774668 (C4 (Web Text))

Max Activating Token Index: 897

Text #19

Max Range: 0.8558. Min Range: -0.8558

Max Act: 0.6047. Min Act: -0.0002

Data Index: 805440 (C4 (Web Text))

Max Activating Token Index: 228