Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 4239 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 3.1797. Min Range: -3.1797

Max Act: 3.1797. Min Act: -0.0000

Data Index: 1117350 (C4 (Web Text))

Max Activating Token Index: 234

Text #1

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.1170. Min Act: -0.0000

Data Index: 238112 (C4 (Web Text))

Max Activating Token Index: 348

Text #2

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.1600. Min Act: -0.0000

Data Index: 24137 (C4 (Web Text))

Max Activating Token Index: 310

Text #3

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.2369. Min Act: -0.0000

Data Index: 114378 (C4 (Web Text))

Max Activating Token Index: 1015

Text #4

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.0129. Min Act: -0.0000

Data Index: 880397 (C4 (Web Text))

Max Activating Token Index: 89

Text #5

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.1112. Min Act: -0.0000

Data Index: 1118766 (C4 (Web Text))

Max Activating Token Index: 765

Text #6

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.0858. Min Act: -0.0000

Data Index: 468535 (C4 (Web Text))

Max Activating Token Index: 235

Text #7

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.9410. Min Act: -0.0000

Data Index: 220893 (C4 (Web Text))

Max Activating Token Index: 923

Text #8

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.0068. Min Act: -0.0000

Data Index: 1142166 (C4 (Web Text))

Max Activating Token Index: 145

Text #9

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.0006. Min Act: -0.0000

Data Index: 970762 (C4 (Web Text))

Max Activating Token Index: 44

Text #10

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.9962. Min Act: -0.0000

Data Index: 809256 (C4 (Web Text))

Max Activating Token Index: 545

Text #11

Max Range: 3.1797. Min Range: -3.1797

Max Act: 2.2299. Min Act: -0.0000

Data Index: 531307 (C4 (Web Text))

Max Activating Token Index: 130

Text #12

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.9695. Min Act: -0.0000

Data Index: 29887 (C4 (Web Text))

Max Activating Token Index: 517

Text #13

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.9178. Min Act: -0.0000

Data Index: 503958 (C4 (Web Text))

Max Activating Token Index: 834

Text #14

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.7660. Min Act: -0.0000

Data Index: 1398164 (Python Code)

Max Activating Token Index: 621

Text #15

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.6671. Min Act: -0.0000

Data Index: 597769 (C4 (Web Text))

Max Activating Token Index: 77

Text #16

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.6446. Min Act: -0.0000

Data Index: 900895 (C4 (Web Text))

Max Activating Token Index: 508

Text #17

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.7498. Min Act: -0.0000

Data Index: 353953 (C4 (Web Text))

Max Activating Token Index: 104

Text #18

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.3794. Min Act: -0.0000

Data Index: 1196341 (C4 (Web Text))

Max Activating Token Index: 278

Text #19

Max Range: 3.1797. Min Range: -3.1797

Max Act: 1.4020. Min Act: -0.0000

Data Index: 340051 (C4 (Web Text))

Max Activating Token Index: 980