Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 640 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0980. Min Act: -0.0002

Data Index: 579058 (C4 (Web Text))

Max Activating Token Index: 171

Text #1

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0885. Min Act: -0.0002

Data Index: 890900 (C4 (Web Text))

Max Activating Token Index: 548

Text #2

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0847. Min Act: -0.0002

Data Index: 767180 (C4 (Web Text))

Max Activating Token Index: 422

Text #3

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0862. Min Act: -0.0001

Data Index: 1317611 (C4 (Web Text))

Max Activating Token Index: 442

Text #4

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0871. Min Act: -0.0002

Data Index: 930027 (C4 (Web Text))

Max Activating Token Index: 425

Text #5

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0835. Min Act: -0.0001

Data Index: 964996 (C4 (Web Text))

Max Activating Token Index: 17

Text #6

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0807. Min Act: -0.0002

Data Index: 390647 (C4 (Web Text))

Max Activating Token Index: 434

Text #7

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0796. Min Act: -0.0001

Data Index: 416375 (C4 (Web Text))

Max Activating Token Index: 604

Text #8

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0768. Min Act: -0.0001

Data Index: 1087775 (C4 (Web Text))

Max Activating Token Index: 670

Text #9

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0782. Min Act: -0.0001

Data Index: 1130016 (C4 (Web Text))

Max Activating Token Index: 186

Text #10

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0760. Min Act: -0.0001

Data Index: 557540 (C4 (Web Text))

Max Activating Token Index: 286

Text #11

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0721. Min Act: -0.0002

Data Index: 1063665 (C4 (Web Text))

Max Activating Token Index: 671

Text #12

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0751. Min Act: -0.0002

Data Index: 886361 (C4 (Web Text))

Max Activating Token Index: 467

Text #13

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0755. Min Act: -0.0001

Data Index: 190009 (C4 (Web Text))

Max Activating Token Index: 240

Text #14

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0742. Min Act: -0.0001

Data Index: 673837 (C4 (Web Text))

Max Activating Token Index: 228

Text #15

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0755. Min Act: -0.0001

Data Index: 225627 (C4 (Web Text))

Max Activating Token Index: 88

Text #16

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0717. Min Act: -0.0001

Data Index: 454639 (C4 (Web Text))

Max Activating Token Index: 953

Text #17

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0713. Min Act: -0.0002

Data Index: 789831 (C4 (Web Text))

Max Activating Token Index: 660

Text #18

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0750. Min Act: -0.0001

Data Index: 136468 (C4 (Web Text))

Max Activating Token Index: 385

Text #19

Max Range: 0.0980. Min Range: -0.0980

Max Act: 0.0715. Min Act: -0.0001

Data Index: 1192609 (C4 (Web Text))

Max Activating Token Index: 922