Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 4218 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0753. Min Act: -0.0002

Data Index: 1612048 (Python Code)

Max Activating Token Index: 407

Text #1

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0606. Min Act: -0.0003

Data Index: 1390812 (Python Code)

Max Activating Token Index: 30

Text #2

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0558. Min Act: -0.0003

Data Index: 704777 (C4 (Web Text))

Max Activating Token Index: 14

Text #3

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0565. Min Act: -0.0002

Data Index: 935391 (C4 (Web Text))

Max Activating Token Index: 14

Text #4

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0538. Min Act: -0.0002

Data Index: 1095316 (C4 (Web Text))

Max Activating Token Index: 11

Text #5

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0502. Min Act: -0.0002

Data Index: 1077139 (C4 (Web Text))

Max Activating Token Index: 23

Text #6

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0483. Min Act: -0.0002

Data Index: 592855 (C4 (Web Text))

Max Activating Token Index: 13

Text #7

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0469. Min Act: -0.0002

Data Index: 314641 (C4 (Web Text))

Max Activating Token Index: 6

Text #8

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0435. Min Act: -0.0002

Data Index: 1338840 (C4 (Web Text))

Max Activating Token Index: 22

Text #9

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0429. Min Act: -0.0002

Data Index: 816660 (C4 (Web Text))

Max Activating Token Index: 40

Text #10

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0415. Min Act: -0.0002

Data Index: 1220139 (C4 (Web Text))

Max Activating Token Index: 8

Text #11

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0433. Min Act: -0.0002

Data Index: 1131827 (C4 (Web Text))

Max Activating Token Index: 39

Text #12

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0428. Min Act: -0.0003

Data Index: 1574736 (Python Code)

Max Activating Token Index: 214

Text #13

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0415. Min Act: -0.0002

Data Index: 607760 (C4 (Web Text))

Max Activating Token Index: 36

Text #14

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0379. Min Act: -0.0002

Data Index: 475816 (C4 (Web Text))

Max Activating Token Index: 671

Text #15

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0363. Min Act: -0.0002

Data Index: 1336984 (C4 (Web Text))

Max Activating Token Index: 46

Text #16

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0384. Min Act: -0.0002

Data Index: 863871 (C4 (Web Text))

Max Activating Token Index: 56

Text #17

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0372. Min Act: -0.0002

Data Index: 1060869 (C4 (Web Text))

Max Activating Token Index: 22

Text #18

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0346. Min Act: -0.0002

Data Index: 582377 (C4 (Web Text))

Max Activating Token Index: 468

Text #19

Max Range: 0.0753. Min Range: -0.0753

Max Act: 0.0344. Min Act: -0.0002

Data Index: 958464 (C4 (Web Text))

Max Activating Token Index: 226