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

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

Neuron 4802 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1516. Min Act: -0.0001

Data Index: 838165 (C4 (Web Text))

Max Activating Token Index: 48

Text #1

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1295. Min Act: -0.0001

Data Index: 1344031 (C4 (Web Text))

Max Activating Token Index: 250

Text #2

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1406. Min Act: -0.0001

Data Index: 727158 (C4 (Web Text))

Max Activating Token Index: 129

Text #3

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1323. Min Act: -0.0001

Data Index: 1228479 (C4 (Web Text))

Max Activating Token Index: 34

Text #4

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1380. Min Act: -0.0001

Data Index: 1479865 (Python Code)

Max Activating Token Index: 190

Text #5

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1256. Min Act: -0.0001

Data Index: 1408604 (Python Code)

Max Activating Token Index: 204

Text #6

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1295. Min Act: -0.0001

Data Index: 899937 (C4 (Web Text))

Max Activating Token Index: 946

Text #7

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1174. Min Act: -0.0001

Data Index: 1423701 (Python Code)

Max Activating Token Index: 521

Text #8

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1259. Min Act: -0.0001

Data Index: 889349 (C4 (Web Text))

Max Activating Token Index: 363

Text #9

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1220. Min Act: -0.0001

Data Index: 39350 (C4 (Web Text))

Max Activating Token Index: 723

Text #10

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1123. Min Act: -0.0001

Data Index: 1482829 (Python Code)

Max Activating Token Index: 1009

Text #11

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1054. Min Act: -0.0001

Data Index: 206402 (C4 (Web Text))

Max Activating Token Index: 383

Text #12

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1035. Min Act: -0.0001

Data Index: 565746 (C4 (Web Text))

Max Activating Token Index: 564

Text #13

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.0997. Min Act: -0.0001

Data Index: 553928 (C4 (Web Text))

Max Activating Token Index: 816

Text #14

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1013. Min Act: -0.0001

Data Index: 1187270 (C4 (Web Text))

Max Activating Token Index: 846

Text #15

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1064. Min Act: -0.0001

Data Index: 655032 (C4 (Web Text))

Max Activating Token Index: 73

Text #16

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.0943. Min Act: -0.0001

Data Index: 251262 (C4 (Web Text))

Max Activating Token Index: 692

Text #17

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1019. Min Act: -0.0001

Data Index: 33754 (C4 (Web Text))

Max Activating Token Index: 1010

Text #18

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.0986. Min Act: -0.0001

Data Index: 1608384 (Python Code)

Max Activating Token Index: 264

Text #19

Max Range: 0.1516. Min Range: -0.1516

Max Act: 0.1049. Min Act: -0.0001

Data Index: 93730 (C4 (Web Text))

Max Activating Token Index: 482