Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 1065 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0783. Min Act: -0.0003

Data Index: 567026 (C4 (Web Text))

Max Activating Token Index: 801

Text #1

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0756. Min Act: -0.0003

Data Index: 445142 (C4 (Web Text))

Max Activating Token Index: 810

Text #2

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0760. Min Act: -0.0004

Data Index: 539651 (C4 (Web Text))

Max Activating Token Index: 760

Text #3

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0755. Min Act: -0.0004

Data Index: 405143 (C4 (Web Text))

Max Activating Token Index: 841

Text #4

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0696. Min Act: -0.0003

Data Index: 958373 (C4 (Web Text))

Max Activating Token Index: 977

Text #5

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0692. Min Act: -0.0002

Data Index: 1040953 (C4 (Web Text))

Max Activating Token Index: 687

Text #6

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0706. Min Act: -0.0003

Data Index: 301164 (C4 (Web Text))

Max Activating Token Index: 747

Text #7

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0682. Min Act: -0.0004

Data Index: 423930 (C4 (Web Text))

Max Activating Token Index: 425

Text #8

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0692. Min Act: -0.0003

Data Index: 696079 (C4 (Web Text))

Max Activating Token Index: 867

Text #9

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0699. Min Act: -0.0003

Data Index: 1111607 (C4 (Web Text))

Max Activating Token Index: 949

Text #10

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0688. Min Act: -0.0003

Data Index: 1088731 (C4 (Web Text))

Max Activating Token Index: 586

Text #11

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0649. Min Act: -0.0003

Data Index: 1285494 (C4 (Web Text))

Max Activating Token Index: 765

Text #12

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0668. Min Act: -0.0003

Data Index: 1103761 (C4 (Web Text))

Max Activating Token Index: 558

Text #13

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0683. Min Act: -0.0003

Data Index: 692636 (C4 (Web Text))

Max Activating Token Index: 406

Text #14

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0643. Min Act: -0.0003

Data Index: 1040010 (C4 (Web Text))

Max Activating Token Index: 880

Text #15

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0666. Min Act: -0.0003

Data Index: 380312 (C4 (Web Text))

Max Activating Token Index: 471

Text #16

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0635. Min Act: -0.0003

Data Index: 951883 (C4 (Web Text))

Max Activating Token Index: 121

Text #17

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0632. Min Act: -0.0004

Data Index: 920647 (C4 (Web Text))

Max Activating Token Index: 377

Text #18

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0648. Min Act: -0.0004

Data Index: 273964 (C4 (Web Text))

Max Activating Token Index: 975

Text #19

Max Range: 0.0783. Min Range: -0.0783

Max Act: 0.0616. Min Act: -0.0003

Data Index: 1281586 (C4 (Web Text))

Max Activating Token Index: 103