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

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

Neuron 4066 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.4325. Min Act: -0.0001

Data Index: 1653040 (Python Code)

Max Activating Token Index: 158

Text #1

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.3465. Min Act: -0.0001

Data Index: 914531 (C4 (Web Text))

Max Activating Token Index: 2

Text #2

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.3422. Min Act: -0.0001

Data Index: 1623925 (Python Code)

Max Activating Token Index: 545

Text #3

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.3416. Min Act: -0.0001

Data Index: 1097463 (C4 (Web Text))

Max Activating Token Index: 30

Text #4

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1667. Min Act: -0.0001

Data Index: 804574 (C4 (Web Text))

Max Activating Token Index: 283

Text #5

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1591. Min Act: -0.0001

Data Index: 1639175 (Python Code)

Max Activating Token Index: 118

Text #6

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.3597. Min Act: -0.0001

Data Index: 1086386 (C4 (Web Text))

Max Activating Token Index: 866

Text #7

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.3417. Min Act: -0.0001

Data Index: 1626288 (Python Code)

Max Activating Token Index: 193

Text #8

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1427. Min Act: -0.0001

Data Index: 253557 (C4 (Web Text))

Max Activating Token Index: 1005

Text #9

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1962. Min Act: -0.0001

Data Index: 1426984 (Python Code)

Max Activating Token Index: 325

Text #10

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.0869. Min Act: -0.0000

Data Index: 1416734 (Python Code)

Max Activating Token Index: 932

Text #11

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1072. Min Act: -0.0000

Data Index: 1399764 (Python Code)

Max Activating Token Index: 971

Text #12

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1940. Min Act: -0.0000

Data Index: 1625347 (Python Code)

Max Activating Token Index: 258

Text #13

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1349. Min Act: -0.0001

Data Index: 644471 (C4 (Web Text))

Max Activating Token Index: 32

Text #14

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1360. Min Act: -0.0000

Data Index: 1571677 (Python Code)

Max Activating Token Index: 625

Text #15

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1293. Min Act: -0.0001

Data Index: 842024 (C4 (Web Text))

Max Activating Token Index: 209

Text #16

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.0590. Min Act: -0.0001

Data Index: 709797 (C4 (Web Text))

Max Activating Token Index: 72

Text #17

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1328. Min Act: -0.0001

Data Index: 764229 (C4 (Web Text))

Max Activating Token Index: 605

Text #18

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.1038. Min Act: -0.0001

Data Index: 95888 (C4 (Web Text))

Max Activating Token Index: 499

Text #19

Max Range: 1.4325. Min Range: -1.4325

Max Act: 1.2237. Min Act: -0.0001

Data Index: 1150694 (C4 (Web Text))

Max Activating Token Index: 1017