Model: SoLU Model: 3 Layers, 2048 Neurons per Layer

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

Neuron 174 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-3l')

Text #0

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0250. Min Act: -0.0006

Data Index: 785920 (C4 (Web Text))

Max Activating Token Index: 815

Text #1

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0216. Min Act: -0.0006

Data Index: 1494227 (Python Code)

Max Activating Token Index: 22

Text #2

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0204. Min Act: -0.0006

Data Index: 156488 (C4 (Web Text))

Max Activating Token Index: 371

Text #3

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0211. Min Act: -0.0006

Data Index: 1299894 (C4 (Web Text))

Max Activating Token Index: 97

Text #4

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0210. Min Act: -0.0006

Data Index: 1386214 (Python Code)

Max Activating Token Index: 402

Text #5

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0199. Min Act: -0.0006

Data Index: 190930 (C4 (Web Text))

Max Activating Token Index: 44

Text #6

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0207. Min Act: -0.0006

Data Index: 999070 (C4 (Web Text))

Max Activating Token Index: 733

Text #7

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0197. Min Act: -0.0006

Data Index: 766949 (C4 (Web Text))

Max Activating Token Index: 1023

Text #8

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0196. Min Act: -0.0006

Data Index: 928020 (C4 (Web Text))

Max Activating Token Index: 527

Text #9

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0201. Min Act: -0.0006

Data Index: 254259 (C4 (Web Text))

Max Activating Token Index: 537

Text #10

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0195. Min Act: -0.0006

Data Index: 1429091 (Python Code)

Max Activating Token Index: 185

Text #11

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0204. Min Act: -0.0006

Data Index: 449043 (C4 (Web Text))

Max Activating Token Index: 702

Text #12

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0198. Min Act: -0.0006

Data Index: 924776 (C4 (Web Text))

Max Activating Token Index: 690

Text #13

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0190. Min Act: -0.0006

Data Index: 1589907 (Python Code)

Max Activating Token Index: 612

Text #14

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0193. Min Act: -0.0006

Data Index: 151593 (C4 (Web Text))

Max Activating Token Index: 30

Text #15

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0192. Min Act: -0.0006

Data Index: 1543924 (Python Code)

Max Activating Token Index: 378

Text #16

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0192. Min Act: -0.0006

Data Index: 792030 (C4 (Web Text))

Max Activating Token Index: 1023

Text #17

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0191. Min Act: -0.0006

Data Index: 695493 (C4 (Web Text))

Max Activating Token Index: 698

Text #18

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0186. Min Act: -0.0006

Data Index: 1450458 (Python Code)

Max Activating Token Index: 1023

Text #19

Max Range: 0.0250. Min Range: -0.0250

Max Act: 0.0188. Min Act: -0.0006

Data Index: 1009497 (C4 (Web Text))

Max Activating Token Index: 1023