Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 110 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1944. Min Act: -0.0001

Data Index: 193339 (C4 (Web Text))

Max Activating Token Index: 837

Text #1

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1837. Min Act: -0.0001

Data Index: 283573 (C4 (Web Text))

Max Activating Token Index: 421

Text #2

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1489. Min Act: -0.0001

Data Index: 1095958 (C4 (Web Text))

Max Activating Token Index: 260

Text #3

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1393. Min Act: -0.0001

Data Index: 391970 (C4 (Web Text))

Max Activating Token Index: 336

Text #4

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1369. Min Act: -0.0001

Data Index: 262709 (C4 (Web Text))

Max Activating Token Index: 653

Text #5

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1274. Min Act: -0.0001

Data Index: 1005947 (C4 (Web Text))

Max Activating Token Index: 394

Text #6

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1317. Min Act: -0.0001

Data Index: 935405 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1322. Min Act: -0.0001

Data Index: 1595308 (Python Code)

Max Activating Token Index: 13

Text #8

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1383. Min Act: -0.0001

Data Index: 948069 (C4 (Web Text))

Max Activating Token Index: 844

Text #9

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1219. Min Act: -0.0001

Data Index: 891764 (C4 (Web Text))

Max Activating Token Index: 216

Text #10

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1267. Min Act: -0.0001

Data Index: 258304 (C4 (Web Text))

Max Activating Token Index: 237

Text #11

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1208. Min Act: -0.0001

Data Index: 682672 (C4 (Web Text))

Max Activating Token Index: 254

Text #12

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1228. Min Act: -0.0001

Data Index: 1629824 (Python Code)

Max Activating Token Index: 18

Text #13

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1183. Min Act: -0.0001

Data Index: 1046983 (C4 (Web Text))

Max Activating Token Index: 386

Text #14

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1106. Min Act: -0.0001

Data Index: 1371446 (Python Code)

Max Activating Token Index: 889

Text #15

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1125. Min Act: -0.0001

Data Index: 519539 (C4 (Web Text))

Max Activating Token Index: 42

Text #16

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1126. Min Act: -0.0001

Data Index: 731021 (C4 (Web Text))

Max Activating Token Index: 194

Text #17

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1093. Min Act: -0.0001

Data Index: 408174 (C4 (Web Text))

Max Activating Token Index: 513

Text #18

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1089. Min Act: -0.0001

Data Index: 649675 (C4 (Web Text))

Max Activating Token Index: 280

Text #19

Max Range: 0.1944. Min Range: -0.1944

Max Act: 0.1115. Min Act: -0.0001

Data Index: 1197688 (C4 (Web Text))

Max Activating Token Index: 773