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

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

Neuron 1478 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2694. Min Act: -0.0003

Data Index: 646257 (C4 (Web Text))

Max Activating Token Index: 337

Text #1

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2593. Min Act: -0.0003

Data Index: 95955 (C4 (Web Text))

Max Activating Token Index: 205

Text #2

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2470. Min Act: -0.0003

Data Index: 8223 (C4 (Web Text))

Max Activating Token Index: 524

Text #3

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2402. Min Act: -0.0003

Data Index: 271017 (C4 (Web Text))

Max Activating Token Index: 118

Text #4

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2301. Min Act: -0.0003

Data Index: 194905 (C4 (Web Text))

Max Activating Token Index: 258

Text #5

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2320. Min Act: -0.0003

Data Index: 626289 (C4 (Web Text))

Max Activating Token Index: 440

Text #6

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2198. Min Act: -0.0003

Data Index: 945800 (C4 (Web Text))

Max Activating Token Index: 98

Text #7

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2245. Min Act: -0.0002

Data Index: 1127160 (C4 (Web Text))

Max Activating Token Index: 409

Text #8

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2241. Min Act: -0.0003

Data Index: 1197694 (C4 (Web Text))

Max Activating Token Index: 768

Text #9

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2233. Min Act: -0.0002

Data Index: 404761 (C4 (Web Text))

Max Activating Token Index: 379

Text #10

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2205. Min Act: -0.0003

Data Index: 793935 (C4 (Web Text))

Max Activating Token Index: 388

Text #11

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2165. Min Act: -0.0003

Data Index: 248459 (C4 (Web Text))

Max Activating Token Index: 569

Text #12

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2162. Min Act: -0.0003

Data Index: 200833 (C4 (Web Text))

Max Activating Token Index: 494

Text #13

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2142. Min Act: -0.0003

Data Index: 261072 (C4 (Web Text))

Max Activating Token Index: 700

Text #14

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2140. Min Act: -0.0003

Data Index: 974273 (C4 (Web Text))

Max Activating Token Index: 536

Text #15

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2133. Min Act: -0.0003

Data Index: 144862 (C4 (Web Text))

Max Activating Token Index: 243

Text #16

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2196. Min Act: -0.0003

Data Index: 1149308 (C4 (Web Text))

Max Activating Token Index: 684

Text #17

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2113. Min Act: -0.0002

Data Index: 1373402 (Python Code)

Max Activating Token Index: 854

Text #18

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2109. Min Act: -0.0003

Data Index: 733110 (C4 (Web Text))

Max Activating Token Index: 768

Text #19

Max Range: 0.2694. Min Range: -0.2694

Max Act: 0.2104. Min Act: -0.0002

Data Index: 627428 (C4 (Web Text))

Max Activating Token Index: 862