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

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

Neuron 1515 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: 0.0647. Min Range: -0.0647

Max Act: 0.0631. Min Act: -0.0003

Data Index: 1147258 (C4 (Web Text))

Max Activating Token Index: 731

Text #1

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0647. Min Act: -0.0003

Data Index: 815142 (C4 (Web Text))

Max Activating Token Index: 571

Text #2

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0615. Min Act: -0.0003

Data Index: 456645 (C4 (Web Text))

Max Activating Token Index: 918

Text #3

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0615. Min Act: -0.0002

Data Index: 380059 (C4 (Web Text))

Max Activating Token Index: 955

Text #4

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0583. Min Act: -0.0004

Data Index: 1296691 (C4 (Web Text))

Max Activating Token Index: 1018

Text #5

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0556. Min Act: -0.0004

Data Index: 446997 (C4 (Web Text))

Max Activating Token Index: 929

Text #6

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0540. Min Act: -0.0003

Data Index: 1277043 (C4 (Web Text))

Max Activating Token Index: 584

Text #7

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0539. Min Act: -0.0003

Data Index: 60325 (C4 (Web Text))

Max Activating Token Index: 603

Text #8

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0535. Min Act: -0.0003

Data Index: 48630 (C4 (Web Text))

Max Activating Token Index: 593

Text #9

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0512. Min Act: -0.0003

Data Index: 349020 (C4 (Web Text))

Max Activating Token Index: 902

Text #10

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0533. Min Act: -0.0003

Data Index: 1040872 (C4 (Web Text))

Max Activating Token Index: 263

Text #11

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0513. Min Act: -0.0003

Data Index: 50018 (C4 (Web Text))

Max Activating Token Index: 838

Text #12

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0522. Min Act: -0.0003

Data Index: 1269272 (C4 (Web Text))

Max Activating Token Index: 288

Text #13

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0532. Min Act: -0.0003

Data Index: 692532 (C4 (Web Text))

Max Activating Token Index: 335

Text #14

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0488. Min Act: -0.0003

Data Index: 447331 (C4 (Web Text))

Max Activating Token Index: 496

Text #15

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0489. Min Act: -0.0003

Data Index: 51616 (C4 (Web Text))

Max Activating Token Index: 611

Text #16

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0488. Min Act: -0.0003

Data Index: 926230 (C4 (Web Text))

Max Activating Token Index: 164

Text #17

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0465. Min Act: -0.0003

Data Index: 636235 (C4 (Web Text))

Max Activating Token Index: 648

Text #18

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0488. Min Act: -0.0002

Data Index: 465008 (C4 (Web Text))

Max Activating Token Index: 467

Text #19

Max Range: 0.0647. Min Range: -0.0647

Max Act: 0.0464. Min Act: -0.0003

Data Index: 99459 (C4 (Web Text))

Max Activating Token Index: 516