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

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

Neuron 3670 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.5977. Min Act: -0.0001

Data Index: 194967 (C4 (Web Text))

Max Activating Token Index: 350

Text #1

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.4069. Min Act: -0.0001

Data Index: 1274102 (C4 (Web Text))

Max Activating Token Index: 852

Text #2

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.1557. Min Act: -0.0001

Data Index: 1098055 (C4 (Web Text))

Max Activating Token Index: 620

Text #3

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.1271. Min Act: -0.0001

Data Index: 350575 (C4 (Web Text))

Max Activating Token Index: 619

Text #4

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.1251. Min Act: -0.0001

Data Index: 1113355 (C4 (Web Text))

Max Activating Token Index: 68

Text #5

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.1537. Min Act: -0.0001

Data Index: 148151 (C4 (Web Text))

Max Activating Token Index: 51

Text #6

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0839. Min Act: -0.0001

Data Index: 809730 (C4 (Web Text))

Max Activating Token Index: 986

Text #7

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0299. Min Act: -0.0001

Data Index: 479398 (C4 (Web Text))

Max Activating Token Index: 206

Text #8

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0971. Min Act: -0.0001

Data Index: 1161226 (C4 (Web Text))

Max Activating Token Index: 890

Text #9

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0798. Min Act: -0.0001

Data Index: 457108 (C4 (Web Text))

Max Activating Token Index: 498

Text #10

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0675. Min Act: -0.0001

Data Index: 376002 (C4 (Web Text))

Max Activating Token Index: 636

Text #11

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9657. Min Act: -0.0001

Data Index: 379112 (C4 (Web Text))

Max Activating Token Index: 690

Text #12

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0554. Min Act: -0.0001

Data Index: 1272772 (C4 (Web Text))

Max Activating Token Index: 432

Text #13

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9932. Min Act: -0.0001

Data Index: 350906 (C4 (Web Text))

Max Activating Token Index: 451

Text #14

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0512. Min Act: -0.0001

Data Index: 569691 (C4 (Web Text))

Max Activating Token Index: 1017

Text #15

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9909. Min Act: -0.0001

Data Index: 1415933 (Python Code)

Max Activating Token Index: 673

Text #16

Max Range: 1.5977. Min Range: -1.5977

Max Act: 1.0361. Min Act: -0.0001

Data Index: 222081 (C4 (Web Text))

Max Activating Token Index: 159

Text #17

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9656. Min Act: -0.0001

Data Index: 493741 (C4 (Web Text))

Max Activating Token Index: 271

Text #18

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9063. Min Act: -0.0001

Data Index: 1277728 (C4 (Web Text))

Max Activating Token Index: 804

Text #19

Max Range: 1.5977. Min Range: -1.5977

Max Act: 0.9849. Min Act: -0.0001

Data Index: 476616 (C4 (Web Text))

Max Activating Token Index: 518