Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 459 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0962. Min Act: -0.0001

Data Index: 881555 (C4 (Web Text))

Max Activating Token Index: 953

Text #1

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0978. Min Act: -0.0001

Data Index: 24277 (C4 (Web Text))

Max Activating Token Index: 412

Text #2

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0770. Min Act: -0.0001

Data Index: 790860 (C4 (Web Text))

Max Activating Token Index: 682

Text #3

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0777. Min Act: -0.0001

Data Index: 723212 (C4 (Web Text))

Max Activating Token Index: 661

Text #4

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0703. Min Act: -0.0001

Data Index: 285574 (C4 (Web Text))

Max Activating Token Index: 748

Text #5

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0748. Min Act: -0.0001

Data Index: 568076 (C4 (Web Text))

Max Activating Token Index: 132

Text #6

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0737. Min Act: -0.0001

Data Index: 43530 (C4 (Web Text))

Max Activating Token Index: 303

Text #7

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0716. Min Act: -0.0001

Data Index: 741969 (C4 (Web Text))

Max Activating Token Index: 721

Text #8

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0791. Min Act: -0.0001

Data Index: 1216067 (C4 (Web Text))

Max Activating Token Index: 367

Text #9

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0698. Min Act: -0.0001

Data Index: 848415 (C4 (Web Text))

Max Activating Token Index: 287

Text #10

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0667. Min Act: -0.0001

Data Index: 1527002 (Python Code)

Max Activating Token Index: 102

Text #11

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0690. Min Act: -0.0001

Data Index: 345454 (C4 (Web Text))

Max Activating Token Index: 305

Text #12

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0737. Min Act: -0.0001

Data Index: 1342536 (C4 (Web Text))

Max Activating Token Index: 616

Text #13

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0679. Min Act: -0.0001

Data Index: 1432484 (Python Code)

Max Activating Token Index: 122

Text #14

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0645. Min Act: -0.0001

Data Index: 1280896 (C4 (Web Text))

Max Activating Token Index: 718

Text #15

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0641. Min Act: -0.0001

Data Index: 1060670 (C4 (Web Text))

Max Activating Token Index: 780

Text #16

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0637. Min Act: -0.0001

Data Index: 714081 (C4 (Web Text))

Max Activating Token Index: 289

Text #17

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0553. Min Act: -0.0001

Data Index: 920038 (C4 (Web Text))

Max Activating Token Index: 854

Text #18

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0701. Min Act: -0.0001

Data Index: 1094981 (C4 (Web Text))

Max Activating Token Index: 527

Text #19

Max Range: 0.0978. Min Range: -0.0978

Max Act: 0.0621. Min Act: -0.0001

Data Index: 523299 (C4 (Web Text))

Max Activating Token Index: 539