Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 2578 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.9211. Min Act: -0.0000

Data Index: 133855 (C4 (Web Text))

Max Activating Token Index: 395

Text #1

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.8017. Min Act: -0.0000

Data Index: 813054 (C4 (Web Text))

Max Activating Token Index: 168

Text #2

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.8605. Min Act: -0.0000

Data Index: 683751 (C4 (Web Text))

Max Activating Token Index: 730

Text #3

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.8225. Min Act: -0.0000

Data Index: 724250 (C4 (Web Text))

Max Activating Token Index: 967

Text #4

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.7347. Min Act: -0.0000

Data Index: 389013 (C4 (Web Text))

Max Activating Token Index: 379

Text #5

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.8059. Min Act: -0.0000

Data Index: 1358996 (C4 (Web Text))

Max Activating Token Index: 925

Text #6

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6775. Min Act: -0.0000

Data Index: 1410045 (Python Code)

Max Activating Token Index: 859

Text #7

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.7074. Min Act: -0.0000

Data Index: 409187 (C4 (Web Text))

Max Activating Token Index: 448

Text #8

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.5627. Min Act: -0.0000

Data Index: 1607838 (Python Code)

Max Activating Token Index: 224

Text #9

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6471. Min Act: -0.0000

Data Index: 1521279 (Python Code)

Max Activating Token Index: 365

Text #10

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6069. Min Act: -0.0000

Data Index: 1372333 (Python Code)

Max Activating Token Index: 136

Text #11

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.5310. Min Act: -0.0000

Data Index: 1595070 (Python Code)

Max Activating Token Index: 515

Text #12

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6630. Min Act: -0.0000

Data Index: 1230912 (C4 (Web Text))

Max Activating Token Index: 548

Text #13

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.5215. Min Act: -0.0000

Data Index: 1629045 (Python Code)

Max Activating Token Index: 473

Text #14

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6875. Min Act: -0.0000

Data Index: 1398329 (Python Code)

Max Activating Token Index: 799

Text #15

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6597. Min Act: -0.0000

Data Index: 473641 (C4 (Web Text))

Max Activating Token Index: 89

Text #16

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6124. Min Act: -0.0000

Data Index: 1518525 (Python Code)

Max Activating Token Index: 880

Text #17

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.5864. Min Act: -0.0000

Data Index: 318360 (C4 (Web Text))

Max Activating Token Index: 337

Text #18

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6014. Min Act: -0.0000

Data Index: 346400 (C4 (Web Text))

Max Activating Token Index: 186

Text #19

Max Range: 0.9211. Min Range: -0.9211

Max Act: 0.6297. Min Act: -0.0000

Data Index: 1071408 (C4 (Web Text))

Max Activating Token Index: 177