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

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

Neuron 4090 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.6574. Min Act: -0.0000

Data Index: 717090 (C4 (Web Text))

Max Activating Token Index: 807

Text #1

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.7220. Min Act: -0.0000

Data Index: 487316 (C4 (Web Text))

Max Activating Token Index: 906

Text #2

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.6461. Min Act: -0.0000

Data Index: 1210915 (C4 (Web Text))

Max Activating Token Index: 319

Text #3

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.6070. Min Act: -0.0000

Data Index: 829903 (C4 (Web Text))

Max Activating Token Index: 111

Text #4

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5934. Min Act: -0.0000

Data Index: 1532049 (Python Code)

Max Activating Token Index: 1015

Text #5

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5770. Min Act: -0.0000

Data Index: 192904 (C4 (Web Text))

Max Activating Token Index: 18

Text #6

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5734. Min Act: -0.0000

Data Index: 1479212 (Python Code)

Max Activating Token Index: 1011

Text #7

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.6099. Min Act: -0.0000

Data Index: 859295 (C4 (Web Text))

Max Activating Token Index: 861

Text #8

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5454. Min Act: -0.0000

Data Index: 548914 (C4 (Web Text))

Max Activating Token Index: 682

Text #9

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5233. Min Act: -0.0000

Data Index: 65184 (C4 (Web Text))

Max Activating Token Index: 956

Text #10

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5338. Min Act: -0.0000

Data Index: 1604741 (Python Code)

Max Activating Token Index: 285

Text #11

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5251. Min Act: -0.0000

Data Index: 54190 (C4 (Web Text))

Max Activating Token Index: 278

Text #12

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5088. Min Act: 0.0000

Data Index: 246949 (C4 (Web Text))

Max Activating Token Index: 976

Text #13

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5214. Min Act: -0.0000

Data Index: 1450926 (Python Code)

Max Activating Token Index: 445

Text #14

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5396. Min Act: -0.0000

Data Index: 332483 (C4 (Web Text))

Max Activating Token Index: 623

Text #15

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5348. Min Act: -0.0000

Data Index: 1580613 (Python Code)

Max Activating Token Index: 187

Text #16

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.4692. Min Act: -0.0000

Data Index: 1452575 (Python Code)

Max Activating Token Index: 893

Text #17

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5254. Min Act: -0.0000

Data Index: 1630426 (Python Code)

Max Activating Token Index: 903

Text #18

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5193. Min Act: 0.0001

Data Index: 1406709 (Python Code)

Max Activating Token Index: 842

Text #19

Max Range: 0.7220. Min Range: -0.7220

Max Act: 0.5587. Min Act: -0.0000

Data Index: 354033 (C4 (Web Text))

Max Activating Token Index: 881