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

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

Neuron 4408 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0234. Min Act: -0.0002

Data Index: 1483797 (Python Code)

Max Activating Token Index: 1

Text #1

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0229. Min Act: -0.0002

Data Index: 48273 (C4 (Web Text))

Max Activating Token Index: 2

Text #2

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0208. Min Act: -0.0002

Data Index: 15254 (C4 (Web Text))

Max Activating Token Index: 914

Text #3

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0209. Min Act: -0.0002

Data Index: 799020 (C4 (Web Text))

Max Activating Token Index: 1

Text #4

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0209. Min Act: -0.0002

Data Index: 268142 (C4 (Web Text))

Max Activating Token Index: 1

Text #5

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0201. Min Act: -0.0002

Data Index: 506085 (C4 (Web Text))

Max Activating Token Index: 1

Text #6

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0199. Min Act: -0.0002

Data Index: 441836 (C4 (Web Text))

Max Activating Token Index: 1

Text #7

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0192. Min Act: -0.0002

Data Index: 1460681 (Python Code)

Max Activating Token Index: 1

Text #8

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0180. Min Act: -0.0002

Data Index: 1427218 (Python Code)

Max Activating Token Index: 2

Text #9

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0172. Min Act: -0.0002

Data Index: 870597 (C4 (Web Text))

Max Activating Token Index: 2

Text #10

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0169. Min Act: -0.0002

Data Index: 156389 (C4 (Web Text))

Max Activating Token Index: 2

Text #11

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0166. Min Act: -0.0002

Data Index: 1630106 (Python Code)

Max Activating Token Index: 569

Text #12

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0171. Min Act: -0.0002

Data Index: 1553714 (Python Code)

Max Activating Token Index: 2

Text #13

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0170. Min Act: -0.0002

Data Index: 1326656 (C4 (Web Text))

Max Activating Token Index: 3

Text #14

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0157. Min Act: -0.0002

Data Index: 779983 (C4 (Web Text))

Max Activating Token Index: 160

Text #15

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0153. Min Act: -0.0002

Data Index: 270914 (C4 (Web Text))

Max Activating Token Index: 2

Text #16

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0153. Min Act: -0.0002

Data Index: 1367364 (Python Code)

Max Activating Token Index: 2

Text #17

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0151. Min Act: -0.0002

Data Index: 511516 (C4 (Web Text))

Max Activating Token Index: 185

Text #18

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0151. Min Act: -0.0002

Data Index: 821179 (C4 (Web Text))

Max Activating Token Index: 2

Text #19

Max Range: 0.0234. Min Range: -0.0234

Max Act: 0.0150. Min Act: -0.0002

Data Index: 1491360 (Python Code)

Max Activating Token Index: 2