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

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

Neuron 4594 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1691. Min Range: -1.1691

Max Act: 1.1691. Min Act: -0.0000

Data Index: 998434 (C4 (Web Text))

Max Activating Token Index: 783

Text #1

Max Range: 1.1691. Min Range: -1.1691

Max Act: 1.0453. Min Act: -0.0000

Data Index: 56843 (C4 (Web Text))

Max Activating Token Index: 947

Text #2

Max Range: 1.1691. Min Range: -1.1691

Max Act: 1.0028. Min Act: -0.0000

Data Index: 1379565 (Python Code)

Max Activating Token Index: 748

Text #3

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.8650. Min Act: -0.0000

Data Index: 228118 (C4 (Web Text))

Max Activating Token Index: 904

Text #4

Max Range: 1.1691. Min Range: -1.1691

Max Act: 1.0545. Min Act: -0.0000

Data Index: 661963 (C4 (Web Text))

Max Activating Token Index: 286

Text #5

Max Range: 1.1691. Min Range: -1.1691

Max Act: 1.0007. Min Act: -0.0000

Data Index: 950686 (C4 (Web Text))

Max Activating Token Index: 262

Text #6

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9640. Min Act: -0.0000

Data Index: 1559104 (Python Code)

Max Activating Token Index: 706

Text #7

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9528. Min Act: -0.0000

Data Index: 196468 (C4 (Web Text))

Max Activating Token Index: 461

Text #8

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9199. Min Act: -0.0000

Data Index: 982336 (C4 (Web Text))

Max Activating Token Index: 627

Text #9

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9008. Min Act: -0.0000

Data Index: 1566816 (Python Code)

Max Activating Token Index: 961

Text #10

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9115. Min Act: -0.0000

Data Index: 1430894 (Python Code)

Max Activating Token Index: 705

Text #11

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9621. Min Act: -0.0000

Data Index: 577584 (C4 (Web Text))

Max Activating Token Index: 282

Text #12

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9247. Min Act: -0.0000

Data Index: 1083867 (C4 (Web Text))

Max Activating Token Index: 23

Text #13

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9446. Min Act: -0.0000

Data Index: 1390743 (Python Code)

Max Activating Token Index: 552

Text #14

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9144. Min Act: -0.0000

Data Index: 287447 (C4 (Web Text))

Max Activating Token Index: 66

Text #15

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9096. Min Act: -0.0000

Data Index: 529306 (C4 (Web Text))

Max Activating Token Index: 912

Text #16

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.8792. Min Act: -0.0000

Data Index: 1524041 (Python Code)

Max Activating Token Index: 240

Text #17

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.9107. Min Act: -0.0000

Data Index: 123000 (C4 (Web Text))

Max Activating Token Index: 179

Text #18

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.8823. Min Act: -0.0000

Data Index: 264134 (C4 (Web Text))

Max Activating Token Index: 334

Text #19

Max Range: 1.1691. Min Range: -1.1691

Max Act: 0.8519. Min Act: -0.0000

Data Index: 642184 (C4 (Web Text))

Max Activating Token Index: 244