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

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

Neuron 1307 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: 0.0287. Min Range: -0.0287

Max Act: 0.0271. Min Act: -0.0006

Data Index: 113818 (C4 (Web Text))

Max Activating Token Index: 413

Text #1

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0270. Min Act: -0.0006

Data Index: 971443 (C4 (Web Text))

Max Activating Token Index: 575

Text #2

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0247. Min Act: -0.0006

Data Index: 60980 (C4 (Web Text))

Max Activating Token Index: 559

Text #3

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0233. Min Act: -0.0006

Data Index: 1599841 (Python Code)

Max Activating Token Index: 509

Text #4

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0248. Min Act: -0.0005

Data Index: 1010638 (C4 (Web Text))

Max Activating Token Index: 64

Text #5

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0255. Min Act: -0.0006

Data Index: 1065475 (C4 (Web Text))

Max Activating Token Index: 112

Text #6

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0287. Min Act: -0.0005

Data Index: 1574292 (Python Code)

Max Activating Token Index: 619

Text #7

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0245. Min Act: -0.0006

Data Index: 443249 (C4 (Web Text))

Max Activating Token Index: 104

Text #8

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0214. Min Act: -0.0006

Data Index: 1429560 (Python Code)

Max Activating Token Index: 656

Text #9

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0248. Min Act: -0.0006

Data Index: 1539418 (Python Code)

Max Activating Token Index: 455

Text #10

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0253. Min Act: -0.0006

Data Index: 202966 (C4 (Web Text))

Max Activating Token Index: 109

Text #11

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0227. Min Act: -0.0005

Data Index: 236818 (C4 (Web Text))

Max Activating Token Index: 159

Text #12

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0237. Min Act: -0.0006

Data Index: 1481002 (Python Code)

Max Activating Token Index: 636

Text #13

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0240. Min Act: -0.0006

Data Index: 1580400 (Python Code)

Max Activating Token Index: 24

Text #14

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0246. Min Act: -0.0006

Data Index: 1609901 (Python Code)

Max Activating Token Index: 738

Text #15

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0201. Min Act: -0.0005

Data Index: 768408 (C4 (Web Text))

Max Activating Token Index: 496

Text #16

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0224. Min Act: -0.0006

Data Index: 1572442 (Python Code)

Max Activating Token Index: 702

Text #17

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0241. Min Act: -0.0006

Data Index: 315672 (C4 (Web Text))

Max Activating Token Index: 38

Text #18

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0237. Min Act: -0.0006

Data Index: 1523284 (Python Code)

Max Activating Token Index: 659

Text #19

Max Range: 0.0287. Min Range: -0.0287

Max Act: 0.0245. Min Act: -0.0006

Data Index: 219171 (C4 (Web Text))

Max Activating Token Index: 634