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

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

Neuron 2575 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.0306. Min Range: -3.0306

Max Act: 3.0238. Min Act: -0.0000

Data Index: 513871 (C4 (Web Text))

Max Activating Token Index: 494

Text #1

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.7399. Min Act: -0.0000

Data Index: 1511093 (Python Code)

Max Activating Token Index: 91

Text #2

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.5374. Min Act: -0.0000

Data Index: 1293956 (C4 (Web Text))

Max Activating Token Index: 754

Text #3

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.7914. Min Act: -0.0000

Data Index: 445480 (C4 (Web Text))

Max Activating Token Index: 664

Text #4

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.7984. Min Act: -0.0000

Data Index: 1394989 (Python Code)

Max Activating Token Index: 519

Text #5

Max Range: 3.0306. Min Range: -3.0306

Max Act: 3.0306. Min Act: -0.0000

Data Index: 1399236 (Python Code)

Max Activating Token Index: 918

Text #6

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.5060. Min Act: -0.0000

Data Index: 705487 (C4 (Web Text))

Max Activating Token Index: 643

Text #7

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.7069. Min Act: -0.0000

Data Index: 624041 (C4 (Web Text))

Max Activating Token Index: 366

Text #8

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.2868. Min Act: -0.0000

Data Index: 709972 (C4 (Web Text))

Max Activating Token Index: 118

Text #9

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.4888. Min Act: -0.0000

Data Index: 619848 (C4 (Web Text))

Max Activating Token Index: 657

Text #10

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.4802. Min Act: -0.0000

Data Index: 1461904 (Python Code)

Max Activating Token Index: 171

Text #11

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.4706. Min Act: -0.0000

Data Index: 1073886 (C4 (Web Text))

Max Activating Token Index: 419

Text #12

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.2158. Min Act: -0.0000

Data Index: 1555265 (Python Code)

Max Activating Token Index: 47

Text #13

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.7182. Min Act: -0.0000

Data Index: 1069841 (C4 (Web Text))

Max Activating Token Index: 564

Text #14

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.4501. Min Act: -0.0000

Data Index: 1413140 (Python Code)

Max Activating Token Index: 1002

Text #15

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.3146. Min Act: -0.0000

Data Index: 921130 (C4 (Web Text))

Max Activating Token Index: 21

Text #16

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.4302. Min Act: -0.0000

Data Index: 640225 (C4 (Web Text))

Max Activating Token Index: 999

Text #17

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.2813. Min Act: -0.0000

Data Index: 795726 (C4 (Web Text))

Max Activating Token Index: 746

Text #18

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.3495. Min Act: -0.0000

Data Index: 452291 (C4 (Web Text))

Max Activating Token Index: 988

Text #19

Max Range: 3.0306. Min Range: -3.0306

Max Act: 2.1136. Min Act: -0.0000

Data Index: 389618 (C4 (Web Text))

Max Activating Token Index: 121