Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3587 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0209. Min Act: -0.0003

Data Index: 298025 (C4 (Web Text))

Max Activating Token Index: 539

Text #1

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0208. Min Act: -0.0003

Data Index: 719735 (C4 (Web Text))

Max Activating Token Index: 215

Text #2

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0210. Min Act: -0.0003

Data Index: 918366 (C4 (Web Text))

Max Activating Token Index: 231

Text #3

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0204. Min Act: -0.0003

Data Index: 442192 (C4 (Web Text))

Max Activating Token Index: 19

Text #4

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0209. Min Act: -0.0003

Data Index: 1605126 (Python Code)

Max Activating Token Index: 54

Text #5

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0205. Min Act: -0.0003

Data Index: 1131560 (C4 (Web Text))

Max Activating Token Index: 31

Text #6

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0185. Min Act: -0.0003

Data Index: 708605 (C4 (Web Text))

Max Activating Token Index: 524

Text #7

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0200. Min Act: -0.0003

Data Index: 1367656 (Python Code)

Max Activating Token Index: 553

Text #8

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0197. Min Act: -0.0003

Data Index: 194337 (C4 (Web Text))

Max Activating Token Index: 647

Text #9

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0203. Min Act: -0.0003

Data Index: 1158414 (C4 (Web Text))

Max Activating Token Index: 33

Text #10

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0187. Min Act: -0.0003

Data Index: 861185 (C4 (Web Text))

Max Activating Token Index: 450

Text #11

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0199. Min Act: -0.0003

Data Index: 1617946 (Python Code)

Max Activating Token Index: 460

Text #12

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0188. Min Act: -0.0003

Data Index: 393756 (C4 (Web Text))

Max Activating Token Index: 393

Text #13

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0203. Min Act: -0.0003

Data Index: 1281653 (C4 (Web Text))

Max Activating Token Index: 72

Text #14

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0195. Min Act: -0.0003

Data Index: 491660 (C4 (Web Text))

Max Activating Token Index: 223

Text #15

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0203. Min Act: -0.0004

Data Index: 796080 (C4 (Web Text))

Max Activating Token Index: 674

Text #16

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0191. Min Act: -0.0003

Data Index: 383176 (C4 (Web Text))

Max Activating Token Index: 17

Text #17

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0193. Min Act: -0.0003

Data Index: 1519390 (Python Code)

Max Activating Token Index: 1023

Text #18

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0197. Min Act: -0.0003

Data Index: 1286104 (C4 (Web Text))

Max Activating Token Index: 39

Text #19

Max Range: 0.0210. Min Range: -0.0210

Max Act: 0.0190. Min Act: -0.0003

Data Index: 178569 (C4 (Web Text))

Max Activating Token Index: 363