Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 659 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0237. Min Act: -0.0006

Data Index: 797869 (C4 (Web Text))

Max Activating Token Index: 1023

Text #1

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0223. Min Act: -0.0006

Data Index: 56992 (C4 (Web Text))

Max Activating Token Index: 1023

Text #2

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0213. Min Act: -0.0006

Data Index: 837636 (C4 (Web Text))

Max Activating Token Index: 492

Text #3

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0222. Min Act: -0.0006

Data Index: 587573 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0181. Min Act: -0.0006

Data Index: 394134 (C4 (Web Text))

Max Activating Token Index: 422

Text #5

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0193. Min Act: -0.0006

Data Index: 908471 (C4 (Web Text))

Max Activating Token Index: 392

Text #6

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0200. Min Act: -0.0006

Data Index: 661977 (C4 (Web Text))

Max Activating Token Index: 1023

Text #7

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0206. Min Act: -0.0006

Data Index: 1562595 (Python Code)

Max Activating Token Index: 592

Text #8

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0155. Min Act: -0.0006

Data Index: 276349 (C4 (Web Text))

Max Activating Token Index: 698

Text #9

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0179. Min Act: -0.0006

Data Index: 602942 (C4 (Web Text))

Max Activating Token Index: 1023

Text #10

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0186. Min Act: -0.0006

Data Index: 192241 (C4 (Web Text))

Max Activating Token Index: 1023

Text #11

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0188. Min Act: -0.0006

Data Index: 237111 (C4 (Web Text))

Max Activating Token Index: 1023

Text #12

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0182. Min Act: -0.0006

Data Index: 1513183 (Python Code)

Max Activating Token Index: 1023

Text #13

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0186. Min Act: -0.0006

Data Index: 558976 (C4 (Web Text))

Max Activating Token Index: 1023

Text #14

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0177. Min Act: -0.0006

Data Index: 45013 (C4 (Web Text))

Max Activating Token Index: 1023

Text #15

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0175. Min Act: -0.0006

Data Index: 1170737 (C4 (Web Text))

Max Activating Token Index: 700

Text #16

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0175. Min Act: -0.0006

Data Index: 59710 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0183. Min Act: -0.0006

Data Index: 1277184 (C4 (Web Text))

Max Activating Token Index: 1023

Text #18

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0190. Min Act: -0.0006

Data Index: 162222 (C4 (Web Text))

Max Activating Token Index: 525

Text #19

Max Range: 0.0237. Min Range: -0.0237

Max Act: 0.0161. Min Act: -0.0006

Data Index: 553497 (C4 (Web Text))

Max Activating Token Index: 270