Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 1028 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0817. Min Act: -0.0001

Data Index: 208780 (C4 (Web Text))

Max Activating Token Index: 286

Text #1

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0751. Min Act: -0.0001

Data Index: 728600 (C4 (Web Text))

Max Activating Token Index: 973

Text #2

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0740. Min Act: -0.0001

Data Index: 632597 (C4 (Web Text))

Max Activating Token Index: 167

Text #3

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0729. Min Act: -0.0001

Data Index: 1051741 (C4 (Web Text))

Max Activating Token Index: 202

Text #4

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0742. Min Act: -0.0001

Data Index: 1001707 (C4 (Web Text))

Max Activating Token Index: 167

Text #5

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0685. Min Act: -0.0001

Data Index: 1055656 (C4 (Web Text))

Max Activating Token Index: 499

Text #6

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0698. Min Act: -0.0001

Data Index: 10510 (C4 (Web Text))

Max Activating Token Index: 533

Text #7

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0701. Min Act: -0.0001

Data Index: 26731 (C4 (Web Text))

Max Activating Token Index: 942

Text #8

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0718. Min Act: -0.0001

Data Index: 15177 (C4 (Web Text))

Max Activating Token Index: 316

Text #9

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0713. Min Act: -0.0001

Data Index: 1082847 (C4 (Web Text))

Max Activating Token Index: 165

Text #10

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0669. Min Act: -0.0001

Data Index: 902676 (C4 (Web Text))

Max Activating Token Index: 160

Text #11

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0668. Min Act: -0.0001

Data Index: 112109 (C4 (Web Text))

Max Activating Token Index: 165

Text #12

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0698. Min Act: -0.0001

Data Index: 1275722 (C4 (Web Text))

Max Activating Token Index: 179

Text #13

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0649. Min Act: -0.0001

Data Index: 394114 (C4 (Web Text))

Max Activating Token Index: 228

Text #14

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0682. Min Act: -0.0001

Data Index: 761268 (C4 (Web Text))

Max Activating Token Index: 970

Text #15

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0619. Min Act: -0.0001

Data Index: 1416813 (Python Code)

Max Activating Token Index: 888

Text #16

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0652. Min Act: -0.0001

Data Index: 354242 (C4 (Web Text))

Max Activating Token Index: 241

Text #17

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0678. Min Act: -0.0001

Data Index: 1330765 (C4 (Web Text))

Max Activating Token Index: 726

Text #18

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0644. Min Act: -0.0001

Data Index: 433947 (C4 (Web Text))

Max Activating Token Index: 88

Text #19

Max Range: 0.0817. Min Range: -0.0817

Max Act: 0.0656. Min Act: -0.0001

Data Index: 296240 (C4 (Web Text))

Max Activating Token Index: 739