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

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

Neuron 2737 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.5024. Min Act: -0.0001

Data Index: 1624424 (Python Code)

Max Activating Token Index: 130

Text #1

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.2073. Min Act: -0.0000

Data Index: 708572 (C4 (Web Text))

Max Activating Token Index: 685

Text #2

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.1632. Min Act: -0.0001

Data Index: 1524622 (Python Code)

Max Activating Token Index: 126

Text #3

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.2504. Min Act: -0.0001

Data Index: 1647744 (Python Code)

Max Activating Token Index: 264

Text #4

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0035. Min Act: -0.0001

Data Index: 1034425 (C4 (Web Text))

Max Activating Token Index: 347

Text #5

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0021. Min Act: -0.0000

Data Index: 213848 (C4 (Web Text))

Max Activating Token Index: 196

Text #6

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0604. Min Act: -0.0000

Data Index: 914105 (C4 (Web Text))

Max Activating Token Index: 352

Text #7

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0274. Min Act: -0.0000

Data Index: 349471 (C4 (Web Text))

Max Activating Token Index: 1021

Text #8

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0179. Min Act: -0.0000

Data Index: 566064 (C4 (Web Text))

Max Activating Token Index: 700

Text #9

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.1085. Min Act: -0.0001

Data Index: 54764 (C4 (Web Text))

Max Activating Token Index: 568

Text #10

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.9532. Min Act: -0.0000

Data Index: 1079244 (C4 (Web Text))

Max Activating Token Index: 278

Text #11

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0367. Min Act: -0.0000

Data Index: 767349 (C4 (Web Text))

Max Activating Token Index: 510

Text #12

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0463. Min Act: -0.0000

Data Index: 1638990 (Python Code)

Max Activating Token Index: 168

Text #13

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.9285. Min Act: -0.0001

Data Index: 1457326 (Python Code)

Max Activating Token Index: 1020

Text #14

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.8118. Min Act: -0.0001

Data Index: 1485745 (Python Code)

Max Activating Token Index: 485

Text #15

Max Range: 2.5024. Min Range: -2.5024

Max Act: 2.0002. Min Act: -0.0000

Data Index: 397700 (C4 (Web Text))

Max Activating Token Index: 84

Text #16

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.8679. Min Act: -0.0000

Data Index: 967397 (C4 (Web Text))

Max Activating Token Index: 933

Text #17

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.7483. Min Act: -0.0001

Data Index: 891029 (C4 (Web Text))

Max Activating Token Index: 802

Text #18

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.7267. Min Act: -0.0001

Data Index: 1380305 (Python Code)

Max Activating Token Index: 586

Text #19

Max Range: 2.5024. Min Range: -2.5024

Max Act: 1.7382. Min Act: -0.0000

Data Index: 377611 (C4 (Web Text))

Max Activating Token Index: 123