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

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

Neuron 618 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.8339. Min Act: -0.0000

Data Index: 758771 (C4 (Web Text))

Max Activating Token Index: 650

Text #1

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.8218. Min Act: -0.0000

Data Index: 877846 (C4 (Web Text))

Max Activating Token Index: 534

Text #2

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.8382. Min Act: -0.0000

Data Index: 934282 (C4 (Web Text))

Max Activating Token Index: 686

Text #3

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.7746. Min Act: -0.0000

Data Index: 591658 (C4 (Web Text))

Max Activating Token Index: 420

Text #4

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.7301. Min Act: -0.0000

Data Index: 73084 (C4 (Web Text))

Max Activating Token Index: 181

Text #5

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.7547. Min Act: -0.0000

Data Index: 92110 (C4 (Web Text))

Max Activating Token Index: 183

Text #6

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.7941. Min Act: -0.0000

Data Index: 1595360 (Python Code)

Max Activating Token Index: 871

Text #7

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6345. Min Act: -0.0000

Data Index: 319463 (C4 (Web Text))

Max Activating Token Index: 34

Text #8

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6572. Min Act: -0.0000

Data Index: 583643 (C4 (Web Text))

Max Activating Token Index: 4

Text #9

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6703. Min Act: -0.0000

Data Index: 1040510 (C4 (Web Text))

Max Activating Token Index: 949

Text #10

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6523. Min Act: -0.0000

Data Index: 940899 (C4 (Web Text))

Max Activating Token Index: 939

Text #11

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6084. Min Act: -0.0000

Data Index: 470855 (C4 (Web Text))

Max Activating Token Index: 714

Text #12

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6249. Min Act: -0.0000

Data Index: 1637961 (Python Code)

Max Activating Token Index: 56

Text #13

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6666. Min Act: -0.0000

Data Index: 24058 (C4 (Web Text))

Max Activating Token Index: 928

Text #14

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.5944. Min Act: -0.0000

Data Index: 580306 (C4 (Web Text))

Max Activating Token Index: 35

Text #15

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.5791. Min Act: -0.0000

Data Index: 730694 (C4 (Web Text))

Max Activating Token Index: 913

Text #16

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.5708. Min Act: -0.0000

Data Index: 236336 (C4 (Web Text))

Max Activating Token Index: 164

Text #17

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6460. Min Act: -0.0000

Data Index: 1514852 (Python Code)

Max Activating Token Index: 277

Text #18

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.6047. Min Act: -0.0000

Data Index: 1175246 (C4 (Web Text))

Max Activating Token Index: 331

Text #19

Max Range: 0.8382. Min Range: -0.8382

Max Act: 0.5583. Min Act: -0.0000

Data Index: 385175 (C4 (Web Text))

Max Activating Token Index: 931