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

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

Neuron 3129 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1761. Min Act: -0.0000

Data Index: 292102 (C4 (Web Text))

Max Activating Token Index: 784

Text #1

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1307. Min Act: -0.0000

Data Index: 689871 (C4 (Web Text))

Max Activating Token Index: 1016

Text #2

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1433. Min Act: -0.0000

Data Index: 352571 (C4 (Web Text))

Max Activating Token Index: 264

Text #3

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1289. Min Act: -0.0000

Data Index: 155346 (C4 (Web Text))

Max Activating Token Index: 271

Text #4

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1251. Min Act: -0.0000

Data Index: 333873 (C4 (Web Text))

Max Activating Token Index: 218

Text #5

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1149. Min Act: -0.0000

Data Index: 1261407 (C4 (Web Text))

Max Activating Token Index: 681

Text #6

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1201. Min Act: -0.0000

Data Index: 296908 (C4 (Web Text))

Max Activating Token Index: 280

Text #7

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1209. Min Act: -0.0000

Data Index: 600189 (C4 (Web Text))

Max Activating Token Index: 315

Text #8

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1089. Min Act: -0.0000

Data Index: 1444184 (Python Code)

Max Activating Token Index: 983

Text #9

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1108. Min Act: -0.0000

Data Index: 293061 (C4 (Web Text))

Max Activating Token Index: 204

Text #10

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1080. Min Act: -0.0000

Data Index: 1050849 (C4 (Web Text))

Max Activating Token Index: 196

Text #11

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1049. Min Act: -0.0000

Data Index: 1057980 (C4 (Web Text))

Max Activating Token Index: 470

Text #12

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1054. Min Act: -0.0000

Data Index: 584755 (C4 (Web Text))

Max Activating Token Index: 702

Text #13

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1034. Min Act: -0.0000

Data Index: 1251436 (C4 (Web Text))

Max Activating Token Index: 38

Text #14

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1066. Min Act: -0.0000

Data Index: 1321596 (C4 (Web Text))

Max Activating Token Index: 804

Text #15

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1071. Min Act: -0.0000

Data Index: 1545161 (Python Code)

Max Activating Token Index: 382

Text #16

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1075. Min Act: -0.0000

Data Index: 359771 (C4 (Web Text))

Max Activating Token Index: 173

Text #17

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0993. Min Act: -0.0000

Data Index: 722024 (C4 (Web Text))

Max Activating Token Index: 533

Text #18

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.1056. Min Act: -0.0000

Data Index: 568851 (C4 (Web Text))

Max Activating Token Index: 445

Text #19

Max Range: 0.1761. Min Range: -0.1761

Max Act: 0.0989. Min Act: -0.0000

Data Index: 1452496 (Python Code)

Max Activating Token Index: 357