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

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

Neuron 4197 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.3069. Min Act: -0.0000

Data Index: 334844 (C4 (Web Text))

Max Activating Token Index: 678

Text #1

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.3002. Min Act: -0.0000

Data Index: 579719 (C4 (Web Text))

Max Activating Token Index: 745

Text #2

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.3032. Min Act: -0.0000

Data Index: 1108591 (C4 (Web Text))

Max Activating Token Index: 941

Text #3

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2843. Min Act: -0.0000

Data Index: 1265672 (C4 (Web Text))

Max Activating Token Index: 847

Text #4

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2902. Min Act: -0.0000

Data Index: 1112413 (C4 (Web Text))

Max Activating Token Index: 940

Text #5

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2651. Min Act: -0.0000

Data Index: 86014 (C4 (Web Text))

Max Activating Token Index: 497

Text #6

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2683. Min Act: -0.0000

Data Index: 787733 (C4 (Web Text))

Max Activating Token Index: 441

Text #7

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2813. Min Act: -0.0000

Data Index: 1177342 (C4 (Web Text))

Max Activating Token Index: 89

Text #8

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2481. Min Act: -0.0000

Data Index: 485071 (C4 (Web Text))

Max Activating Token Index: 731

Text #9

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2472. Min Act: -0.0000

Data Index: 1281937 (C4 (Web Text))

Max Activating Token Index: 734

Text #10

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2537. Min Act: -0.0000

Data Index: 1207141 (C4 (Web Text))

Max Activating Token Index: 723

Text #11

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2429. Min Act: -0.0000

Data Index: 174917 (C4 (Web Text))

Max Activating Token Index: 250

Text #12

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2413. Min Act: -0.0000

Data Index: 532936 (C4 (Web Text))

Max Activating Token Index: 434

Text #13

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2365. Min Act: -0.0000

Data Index: 4682 (C4 (Web Text))

Max Activating Token Index: 671

Text #14

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2343. Min Act: -0.0000

Data Index: 237986 (C4 (Web Text))

Max Activating Token Index: 264

Text #15

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2238. Min Act: -0.0000

Data Index: 1292952 (C4 (Web Text))

Max Activating Token Index: 179

Text #16

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2166. Min Act: -0.0000

Data Index: 253283 (C4 (Web Text))

Max Activating Token Index: 182

Text #17

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2189. Min Act: -0.0000

Data Index: 1529039 (Python Code)

Max Activating Token Index: 221

Text #18

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2290. Min Act: -0.0000

Data Index: 1135814 (C4 (Web Text))

Max Activating Token Index: 863

Text #19

Max Range: 0.3069. Min Range: -0.3069

Max Act: 0.2115. Min Act: -0.0000

Data Index: 52084 (C4 (Web Text))

Max Activating Token Index: 907