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

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

Neuron 4321 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: 0.0247. Min Range: -0.0247

Max Act: 0.0247. Min Act: -0.0003

Data Index: 1111077 (C4 (Web Text))

Max Activating Token Index: 278

Text #1

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0234. Min Act: -0.0003

Data Index: 1462515 (Python Code)

Max Activating Token Index: 480

Text #2

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0233. Min Act: -0.0003

Data Index: 618391 (C4 (Web Text))

Max Activating Token Index: 85

Text #3

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0229. Min Act: -0.0003

Data Index: 653646 (C4 (Web Text))

Max Activating Token Index: 345

Text #4

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0218. Min Act: -0.0003

Data Index: 268930 (C4 (Web Text))

Max Activating Token Index: 21

Text #5

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0225. Min Act: -0.0003

Data Index: 498293 (C4 (Web Text))

Max Activating Token Index: 748

Text #6

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0216. Min Act: -0.0003

Data Index: 1236355 (C4 (Web Text))

Max Activating Token Index: 362

Text #7

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0213. Min Act: -0.0003

Data Index: 388044 (C4 (Web Text))

Max Activating Token Index: 355

Text #8

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0209. Min Act: -0.0003

Data Index: 1021927 (C4 (Web Text))

Max Activating Token Index: 383

Text #9

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0214. Min Act: -0.0003

Data Index: 1594770 (Python Code)

Max Activating Token Index: 175

Text #10

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0202. Min Act: -0.0003

Data Index: 692419 (C4 (Web Text))

Max Activating Token Index: 449

Text #11

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0201. Min Act: -0.0003

Data Index: 1193747 (C4 (Web Text))

Max Activating Token Index: 125

Text #12

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0200. Min Act: -0.0003

Data Index: 1448822 (Python Code)

Max Activating Token Index: 89

Text #13

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0197. Min Act: -0.0003

Data Index: 889271 (C4 (Web Text))

Max Activating Token Index: 882

Text #14

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0190. Min Act: -0.0003

Data Index: 948483 (C4 (Web Text))

Max Activating Token Index: 204

Text #15

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0189. Min Act: -0.0003

Data Index: 38552 (C4 (Web Text))

Max Activating Token Index: 640

Text #16

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0190. Min Act: -0.0003

Data Index: 622703 (C4 (Web Text))

Max Activating Token Index: 316

Text #17

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0191. Min Act: -0.0002

Data Index: 1546609 (Python Code)

Max Activating Token Index: 978

Text #18

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0193. Min Act: -0.0002

Data Index: 1442623 (Python Code)

Max Activating Token Index: 135

Text #19

Max Range: 0.0247. Min Range: -0.0247

Max Act: 0.0186. Min Act: -0.0003

Data Index: 671057 (C4 (Web Text))

Max Activating Token Index: 478