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

Dataset: The Pile

Neuron 2006 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.2161. Min Range: -4.2161

Max Act: 4.0441. Min Act: -0.0000

Data Index: 805696 (The Pile)

Max Activating Token Index: 49

Text #1

Max Range: 4.2161. Min Range: -4.2161

Max Act: 4.0079. Min Act: -0.0000

Data Index: 531960 (The Pile)

Max Activating Token Index: 1002

Text #2

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.7301. Min Act: -0.0000

Data Index: 1894346 (The Pile)

Max Activating Token Index: 51

Text #3

Max Range: 4.2161. Min Range: -4.2161

Max Act: 4.2161. Min Act: -0.0000

Data Index: 496207 (The Pile)

Max Activating Token Index: 603

Text #4

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.8520. Min Act: -0.0000

Data Index: 1957954 (The Pile)

Max Activating Token Index: 624

Text #5

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.8349. Min Act: -0.0000

Data Index: 982573 (The Pile)

Max Activating Token Index: 788

Text #6

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.8213. Min Act: -0.0000

Data Index: 284079 (The Pile)

Max Activating Token Index: 609

Text #7

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.8590. Min Act: -0.0000

Data Index: 1894228 (The Pile)

Max Activating Token Index: 711

Text #8

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.6513. Min Act: -0.0000

Data Index: 233210 (The Pile)

Max Activating Token Index: 201

Text #9

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.5960. Min Act: -0.0000

Data Index: 958764 (The Pile)

Max Activating Token Index: 803

Text #10

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.4137. Min Act: -0.0000

Data Index: 983225 (The Pile)

Max Activating Token Index: 108

Text #11

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.5655. Min Act: -0.0000

Data Index: 794081 (The Pile)

Max Activating Token Index: 825

Text #12

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.4207. Min Act: -0.0000

Data Index: 1764605 (The Pile)

Max Activating Token Index: 992

Text #13

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.4192. Min Act: -0.0000

Data Index: 1721406 (The Pile)

Max Activating Token Index: 136

Text #14

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.5643. Min Act: -0.0000

Data Index: 1603331 (The Pile)

Max Activating Token Index: 677

Text #15

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.5272. Min Act: -0.0000

Data Index: 660816 (The Pile)

Max Activating Token Index: 826

Text #16

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.4789. Min Act: -0.0000

Data Index: 1304826 (The Pile)

Max Activating Token Index: 1012

Text #17

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.3482. Min Act: -0.0000

Data Index: 1914586 (The Pile)

Max Activating Token Index: 704

Text #18

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.6155. Min Act: -0.0000

Data Index: 350569 (The Pile)

Max Activating Token Index: 327

Text #19

Max Range: 4.2161. Min Range: -4.2161

Max Act: 3.4502. Min Act: -0.0000

Data Index: 1973132 (The Pile)

Max Activating Token Index: 980