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

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

Neuron 4564 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: 0.3372. Min Range: -0.3372

Max Act: 0.3372. Min Act: -0.0000

Data Index: 536849 (C4 (Web Text))

Max Activating Token Index: 879

Text #1

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.3010. Min Act: -0.0000

Data Index: 844313 (C4 (Web Text))

Max Activating Token Index: 368

Text #2

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2838. Min Act: -0.0000

Data Index: 191390 (C4 (Web Text))

Max Activating Token Index: 473

Text #3

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2620. Min Act: -0.0000

Data Index: 323958 (C4 (Web Text))

Max Activating Token Index: 633

Text #4

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2419. Min Act: -0.0000

Data Index: 1257990 (C4 (Web Text))

Max Activating Token Index: 493

Text #5

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2487. Min Act: -0.0000

Data Index: 926009 (C4 (Web Text))

Max Activating Token Index: 476

Text #6

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2245. Min Act: -0.0000

Data Index: 1279868 (C4 (Web Text))

Max Activating Token Index: 149

Text #7

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2055. Min Act: -0.0000

Data Index: 293048 (C4 (Web Text))

Max Activating Token Index: 502

Text #8

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.2034. Min Act: -0.0000

Data Index: 568073 (C4 (Web Text))

Max Activating Token Index: 112

Text #9

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1918. Min Act: -0.0000

Data Index: 838847 (C4 (Web Text))

Max Activating Token Index: 570

Text #10

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1884. Min Act: -0.0000

Data Index: 1120846 (C4 (Web Text))

Max Activating Token Index: 413

Text #11

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1944. Min Act: -0.0000

Data Index: 874672 (C4 (Web Text))

Max Activating Token Index: 851

Text #12

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1968. Min Act: -0.0000

Data Index: 294003 (C4 (Web Text))

Max Activating Token Index: 384

Text #13

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1846. Min Act: -0.0000

Data Index: 1412117 (Python Code)

Max Activating Token Index: 33

Text #14

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1873. Min Act: -0.0000

Data Index: 1131073 (C4 (Web Text))

Max Activating Token Index: 302

Text #15

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1835. Min Act: -0.0000

Data Index: 519491 (C4 (Web Text))

Max Activating Token Index: 20

Text #16

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1599. Min Act: -0.0000

Data Index: 499513 (C4 (Web Text))

Max Activating Token Index: 262

Text #17

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1699. Min Act: -0.0000

Data Index: 758867 (C4 (Web Text))

Max Activating Token Index: 122

Text #18

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1677. Min Act: -0.0000

Data Index: 339723 (C4 (Web Text))

Max Activating Token Index: 294

Text #19

Max Range: 0.3372. Min Range: -0.3372

Max Act: 0.1685. Min Act: -0.0000

Data Index: 905171 (C4 (Web Text))

Max Activating Token Index: 709