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

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

Neuron 3108 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: 1.4982. Min Range: -1.4982

Max Act: 1.4982. Min Act: -0.0001

Data Index: 220296 (C4 (Web Text))

Max Activating Token Index: 87

Text #1

Max Range: 1.4982. Min Range: -1.4982

Max Act: 1.3210. Min Act: -0.0001

Data Index: 332270 (C4 (Web Text))

Max Activating Token Index: 769

Text #2

Max Range: 1.4982. Min Range: -1.4982

Max Act: 1.0095. Min Act: -0.0001

Data Index: 942601 (C4 (Web Text))

Max Activating Token Index: 747

Text #3

Max Range: 1.4982. Min Range: -1.4982

Max Act: 1.1346. Min Act: -0.0000

Data Index: 1321846 (C4 (Web Text))

Max Activating Token Index: 900

Text #4

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9891. Min Act: -0.0001

Data Index: 973636 (C4 (Web Text))

Max Activating Token Index: 917

Text #5

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9694. Min Act: -0.0001

Data Index: 402118 (C4 (Web Text))

Max Activating Token Index: 569

Text #6

Max Range: 1.4982. Min Range: -1.4982

Max Act: 1.0247. Min Act: -0.0001

Data Index: 1085235 (C4 (Web Text))

Max Activating Token Index: 846

Text #7

Max Range: 1.4982. Min Range: -1.4982

Max Act: 1.0020. Min Act: -0.0001

Data Index: 1272978 (C4 (Web Text))

Max Activating Token Index: 641

Text #8

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9141. Min Act: -0.0001

Data Index: 543322 (C4 (Web Text))

Max Activating Token Index: 327

Text #9

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9616. Min Act: -0.0001

Data Index: 458226 (C4 (Web Text))

Max Activating Token Index: 182

Text #10

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9240. Min Act: -0.0001

Data Index: 611223 (C4 (Web Text))

Max Activating Token Index: 467

Text #11

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9400. Min Act: -0.0001

Data Index: 903007 (C4 (Web Text))

Max Activating Token Index: 495

Text #12

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9190. Min Act: -0.0001

Data Index: 1022403 (C4 (Web Text))

Max Activating Token Index: 909

Text #13

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8625. Min Act: -0.0001

Data Index: 1133213 (C4 (Web Text))

Max Activating Token Index: 946

Text #14

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.9006. Min Act: -0.0001

Data Index: 147190 (C4 (Web Text))

Max Activating Token Index: 885

Text #15

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8845. Min Act: -0.0001

Data Index: 239216 (C4 (Web Text))

Max Activating Token Index: 647

Text #16

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8224. Min Act: -0.0001

Data Index: 1301231 (C4 (Web Text))

Max Activating Token Index: 173

Text #17

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8972. Min Act: -0.0001

Data Index: 1072375 (C4 (Web Text))

Max Activating Token Index: 571

Text #18

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8070. Min Act: -0.0001

Data Index: 363168 (C4 (Web Text))

Max Activating Token Index: 830

Text #19

Max Range: 1.4982. Min Range: -1.4982

Max Act: 0.8005. Min Act: -0.0001

Data Index: 180081 (C4 (Web Text))

Max Activating Token Index: 822