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

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

Neuron 490 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: 0.0797. Min Range: -0.0797

Max Act: 0.0797. Min Act: -0.0003

Data Index: 514717 (C4 (Web Text))

Max Activating Token Index: 92

Text #1

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0769. Min Act: -0.0003

Data Index: 1382538 (Python Code)

Max Activating Token Index: 25

Text #2

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0736. Min Act: -0.0003

Data Index: 935600 (C4 (Web Text))

Max Activating Token Index: 508

Text #3

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0664. Min Act: -0.0003

Data Index: 15559 (C4 (Web Text))

Max Activating Token Index: 732

Text #4

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0647. Min Act: -0.0003

Data Index: 1232587 (C4 (Web Text))

Max Activating Token Index: 305

Text #5

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0666. Min Act: -0.0003

Data Index: 659280 (C4 (Web Text))

Max Activating Token Index: 282

Text #6

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0598. Min Act: -0.0003

Data Index: 377131 (C4 (Web Text))

Max Activating Token Index: 221

Text #7

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0615. Min Act: -0.0003

Data Index: 1576676 (Python Code)

Max Activating Token Index: 589

Text #8

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0644. Min Act: -0.0003

Data Index: 1209076 (C4 (Web Text))

Max Activating Token Index: 112

Text #9

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0635. Min Act: -0.0003

Data Index: 901241 (C4 (Web Text))

Max Activating Token Index: 296

Text #10

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0640. Min Act: -0.0003

Data Index: 629115 (C4 (Web Text))

Max Activating Token Index: 640

Text #11

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0651. Min Act: -0.0003

Data Index: 1311174 (C4 (Web Text))

Max Activating Token Index: 543

Text #12

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0619. Min Act: -0.0003

Data Index: 1356261 (C4 (Web Text))

Max Activating Token Index: 927

Text #13

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0587. Min Act: -0.0003

Data Index: 628781 (C4 (Web Text))

Max Activating Token Index: 247

Text #14

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0581. Min Act: -0.0003

Data Index: 1008497 (C4 (Web Text))

Max Activating Token Index: 201

Text #15

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0611. Min Act: -0.0003

Data Index: 1247059 (C4 (Web Text))

Max Activating Token Index: 861

Text #16

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0598. Min Act: -0.0003

Data Index: 901796 (C4 (Web Text))

Max Activating Token Index: 78

Text #17

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0565. Min Act: -0.0003

Data Index: 612342 (C4 (Web Text))

Max Activating Token Index: 1007

Text #18

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0590. Min Act: -0.0003

Data Index: 33695 (C4 (Web Text))

Max Activating Token Index: 515

Text #19

Max Range: 0.0797. Min Range: -0.0797

Max Act: 0.0574. Min Act: -0.0003

Data Index: 700968 (C4 (Web Text))

Max Activating Token Index: 412