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

Dataset: The Pile

Neuron 100 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.2918. Min Range: -3.2918

Max Act: 3.2918. Min Act: -0.0000

Data Index: 1227588 (The Pile)

Max Activating Token Index: 280

Text #1

Max Range: 3.2918. Min Range: -3.2918

Max Act: 3.2840. Min Act: -0.0000

Data Index: 1147683 (The Pile)

Max Activating Token Index: 276

Text #2

Max Range: 3.2918. Min Range: -3.2918

Max Act: 2.3526. Min Act: -0.0000

Data Index: 650604 (The Pile)

Max Activating Token Index: 486

Text #3

Max Range: 3.2918. Min Range: -3.2918

Max Act: 2.4042. Min Act: -0.0000

Data Index: 50213 (The Pile)

Max Activating Token Index: 571

Text #4

Max Range: 3.2918. Min Range: -3.2918

Max Act: 2.2060. Min Act: -0.0000

Data Index: 879807 (The Pile)

Max Activating Token Index: 787

Text #5

Max Range: 3.2918. Min Range: -3.2918

Max Act: 2.1830. Min Act: -0.0000

Data Index: 520788 (The Pile)

Max Activating Token Index: 402

Text #6

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.7228. Min Act: -0.0000

Data Index: 989846 (The Pile)

Max Activating Token Index: 832

Text #7

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.7204. Min Act: -0.0000

Data Index: 263635 (The Pile)

Max Activating Token Index: 764

Text #8

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.5927. Min Act: -0.0000

Data Index: 1312868 (The Pile)

Max Activating Token Index: 901

Text #9

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.6261. Min Act: -0.0000

Data Index: 715982 (The Pile)

Max Activating Token Index: 699

Text #10

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.4661. Min Act: -0.0000

Data Index: 867990 (The Pile)

Max Activating Token Index: 217

Text #11

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.4105. Min Act: -0.0000

Data Index: 710395 (The Pile)

Max Activating Token Index: 417

Text #12

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3147. Min Act: -0.0000

Data Index: 1351319 (The Pile)

Max Activating Token Index: 314

Text #13

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3251. Min Act: -0.0000

Data Index: 331621 (The Pile)

Max Activating Token Index: 807

Text #14

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.4504. Min Act: -0.0000

Data Index: 613215 (The Pile)

Max Activating Token Index: 766

Text #15

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3874. Min Act: -0.0000

Data Index: 1624824 (The Pile)

Max Activating Token Index: 334

Text #16

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3021. Min Act: -0.0000

Data Index: 958184 (The Pile)

Max Activating Token Index: 820

Text #17

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3344. Min Act: -0.0000

Data Index: 692741 (The Pile)

Max Activating Token Index: 425

Text #18

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.1131. Min Act: -0.0000

Data Index: 1058434 (The Pile)

Max Activating Token Index: 182

Text #19

Max Range: 3.2918. Min Range: -3.2918

Max Act: 1.3209. Min Act: -0.0000

Data Index: 1507619 (The Pile)

Max Activating Token Index: 82