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

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

Neuron 125 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: 3.0530. Min Range: -3.0530

Max Act: 3.0530. Min Act: -0.0001

Data Index: 443273 (C4 (Web Text))

Max Activating Token Index: 395

Text #1

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.5702. Min Act: -0.0001

Data Index: 878898 (C4 (Web Text))

Max Activating Token Index: 31

Text #2

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.5224. Min Act: -0.0001

Data Index: 98669 (C4 (Web Text))

Max Activating Token Index: 596

Text #3

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.4056. Min Act: -0.0001

Data Index: 367933 (C4 (Web Text))

Max Activating Token Index: 46

Text #4

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.1368. Min Act: -0.0000

Data Index: 119877 (C4 (Web Text))

Max Activating Token Index: 819

Text #5

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3720. Min Act: -0.0001

Data Index: 1130867 (C4 (Web Text))

Max Activating Token Index: 872

Text #6

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3473. Min Act: -0.0000

Data Index: 266996 (C4 (Web Text))

Max Activating Token Index: 68

Text #7

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3530. Min Act: -0.0000

Data Index: 463516 (C4 (Web Text))

Max Activating Token Index: 878

Text #8

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3166. Min Act: -0.0001

Data Index: 468927 (C4 (Web Text))

Max Activating Token Index: 114

Text #9

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3141. Min Act: -0.0001

Data Index: 1256331 (C4 (Web Text))

Max Activating Token Index: 449

Text #10

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3119. Min Act: -0.0001

Data Index: 877634 (C4 (Web Text))

Max Activating Token Index: 72

Text #11

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.1625. Min Act: -0.0001

Data Index: 444103 (C4 (Web Text))

Max Activating Token Index: 853

Text #12

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.2687. Min Act: -0.0000

Data Index: 500191 (C4 (Web Text))

Max Activating Token Index: 732

Text #13

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.0612. Min Act: -0.0000

Data Index: 405735 (C4 (Web Text))

Max Activating Token Index: 668

Text #14

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.1538. Min Act: -0.0000

Data Index: 583290 (C4 (Web Text))

Max Activating Token Index: 731

Text #15

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.1437. Min Act: -0.0001

Data Index: 692227 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.3718. Min Act: -0.0001

Data Index: 451076 (C4 (Web Text))

Max Activating Token Index: 820

Text #17

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.2402. Min Act: -0.0000

Data Index: 333819 (C4 (Web Text))

Max Activating Token Index: 66

Text #18

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.2472. Min Act: -0.0000

Data Index: 299938 (C4 (Web Text))

Max Activating Token Index: 370

Text #19

Max Range: 3.0530. Min Range: -3.0530

Max Act: 2.1206. Min Act: -0.0000

Data Index: 609341 (C4 (Web Text))

Max Activating Token Index: 40