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

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

Neuron 2644 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: 1.2393. Min Range: -1.2393

Max Act: 1.2393. Min Act: -0.0000

Data Index: 564858 (C4 (Web Text))

Max Activating Token Index: 486

Text #1

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.2084. Min Act: -0.0000

Data Index: 1313173 (C4 (Web Text))

Max Activating Token Index: 158

Text #2

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.1439. Min Act: -0.0000

Data Index: 358221 (C4 (Web Text))

Max Activating Token Index: 747

Text #3

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.1323. Min Act: -0.0000

Data Index: 49756 (C4 (Web Text))

Max Activating Token Index: 941

Text #4

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.1243. Min Act: -0.0000

Data Index: 729001 (C4 (Web Text))

Max Activating Token Index: 697

Text #5

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.1101. Min Act: -0.0000

Data Index: 1291821 (C4 (Web Text))

Max Activating Token Index: 94

Text #6

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0471. Min Act: -0.0000

Data Index: 581097 (C4 (Web Text))

Max Activating Token Index: 878

Text #7

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0821. Min Act: -0.0000

Data Index: 1102263 (C4 (Web Text))

Max Activating Token Index: 727

Text #8

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0178. Min Act: -0.0000

Data Index: 311079 (C4 (Web Text))

Max Activating Token Index: 385

Text #9

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9488. Min Act: -0.0000

Data Index: 286583 (C4 (Web Text))

Max Activating Token Index: 789

Text #10

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0324. Min Act: -0.0000

Data Index: 123827 (C4 (Web Text))

Max Activating Token Index: 328

Text #11

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9977. Min Act: -0.0000

Data Index: 1071183 (C4 (Web Text))

Max Activating Token Index: 840

Text #12

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0166. Min Act: -0.0000

Data Index: 166093 (C4 (Web Text))

Max Activating Token Index: 133

Text #13

Max Range: 1.2393. Min Range: -1.2393

Max Act: 1.0103. Min Act: -0.0000

Data Index: 430145 (C4 (Web Text))

Max Activating Token Index: 881

Text #14

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.8936. Min Act: -0.0000

Data Index: 440647 (C4 (Web Text))

Max Activating Token Index: 779

Text #15

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9363. Min Act: -0.0000

Data Index: 900980 (C4 (Web Text))

Max Activating Token Index: 521

Text #16

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9643. Min Act: -0.0000

Data Index: 1111224 (C4 (Web Text))

Max Activating Token Index: 55

Text #17

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9388. Min Act: -0.0000

Data Index: 114911 (C4 (Web Text))

Max Activating Token Index: 286

Text #18

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9201. Min Act: -0.0000

Data Index: 1234080 (C4 (Web Text))

Max Activating Token Index: 381

Text #19

Max Range: 1.2393. Min Range: -1.2393

Max Act: 0.9467. Min Act: -0.0000

Data Index: 1481587 (Python Code)

Max Activating Token Index: 765