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

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

Neuron 1217 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: 0.0662. Min Range: -0.0662

Max Act: 0.0662. Min Act: -0.0002

Data Index: 1422214 (Python Code)

Max Activating Token Index: 499

Text #1

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0598. Min Act: -0.0002

Data Index: 309015 (C4 (Web Text))

Max Activating Token Index: 650

Text #2

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0606. Min Act: -0.0002

Data Index: 742032 (C4 (Web Text))

Max Activating Token Index: 394

Text #3

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0583. Min Act: -0.0002

Data Index: 1490299 (Python Code)

Max Activating Token Index: 44

Text #4

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0574. Min Act: -0.0002

Data Index: 511824 (C4 (Web Text))

Max Activating Token Index: 718

Text #5

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0550. Min Act: -0.0002

Data Index: 821359 (C4 (Web Text))

Max Activating Token Index: 837

Text #6

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0549. Min Act: -0.0002

Data Index: 1126377 (C4 (Web Text))

Max Activating Token Index: 153

Text #7

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0543. Min Act: -0.0002

Data Index: 453155 (C4 (Web Text))

Max Activating Token Index: 375

Text #8

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0576. Min Act: -0.0002

Data Index: 1497475 (Python Code)

Max Activating Token Index: 749

Text #9

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0523. Min Act: -0.0002

Data Index: 65314 (C4 (Web Text))

Max Activating Token Index: 606

Text #10

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0544. Min Act: -0.0002

Data Index: 283388 (C4 (Web Text))

Max Activating Token Index: 50

Text #11

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0545. Min Act: -0.0002

Data Index: 1347614 (C4 (Web Text))

Max Activating Token Index: 765

Text #12

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0561. Min Act: -0.0002

Data Index: 984469 (C4 (Web Text))

Max Activating Token Index: 671

Text #13

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0530. Min Act: -0.0002

Data Index: 1013393 (C4 (Web Text))

Max Activating Token Index: 238

Text #14

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0556. Min Act: -0.0002

Data Index: 1491239 (Python Code)

Max Activating Token Index: 599

Text #15

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0553. Min Act: -0.0002

Data Index: 815990 (C4 (Web Text))

Max Activating Token Index: 606

Text #16

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0530. Min Act: -0.0002

Data Index: 409507 (C4 (Web Text))

Max Activating Token Index: 415

Text #17

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0587. Min Act: -0.0002

Data Index: 940408 (C4 (Web Text))

Max Activating Token Index: 304

Text #18

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0516. Min Act: -0.0001

Data Index: 716975 (C4 (Web Text))

Max Activating Token Index: 514

Text #19

Max Range: 0.0662. Min Range: -0.0662

Max Act: 0.0534. Min Act: -0.0002

Data Index: 190110 (C4 (Web Text))

Max Activating Token Index: 905