Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1987 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0451. Min Act: -0.0001

Data Index: 833057 (C4 (Web Text))

Max Activating Token Index: 211

Text #1

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0353. Min Act: -0.0001

Data Index: 335513 (C4 (Web Text))

Max Activating Token Index: 120

Text #2

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0271. Min Act: -0.0001

Data Index: 384322 (C4 (Web Text))

Max Activating Token Index: 130

Text #3

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0297. Min Act: -0.0001

Data Index: 533831 (C4 (Web Text))

Max Activating Token Index: 321

Text #4

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0315. Min Act: -0.0001

Data Index: 61123 (C4 (Web Text))

Max Activating Token Index: 270

Text #5

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0306. Min Act: -0.0001

Data Index: 712720 (C4 (Web Text))

Max Activating Token Index: 613

Text #6

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0275. Min Act: -0.0001

Data Index: 321446 (C4 (Web Text))

Max Activating Token Index: 78

Text #7

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0278. Min Act: -0.0001

Data Index: 1205126 (C4 (Web Text))

Max Activating Token Index: 840

Text #8

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0307. Min Act: -0.0001

Data Index: 1552414 (Python Code)

Max Activating Token Index: 240

Text #9

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0324. Min Act: -0.0001

Data Index: 57547 (C4 (Web Text))

Max Activating Token Index: 618

Text #10

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0239. Min Act: -0.0001

Data Index: 964369 (C4 (Web Text))

Max Activating Token Index: 343

Text #11

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0291. Min Act: -0.0001

Data Index: 350785 (C4 (Web Text))

Max Activating Token Index: 121

Text #12

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0260. Min Act: -0.0001

Data Index: 397145 (C4 (Web Text))

Max Activating Token Index: 435

Text #13

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0234. Min Act: -0.0001

Data Index: 590380 (C4 (Web Text))

Max Activating Token Index: 699

Text #14

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0283. Min Act: -0.0001

Data Index: 1334728 (C4 (Web Text))

Max Activating Token Index: 819

Text #15

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0252. Min Act: -0.0001

Data Index: 1648532 (Python Code)

Max Activating Token Index: 260

Text #16

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0235. Min Act: -0.0001

Data Index: 1655316 (Python Code)

Max Activating Token Index: 392

Text #17

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0256. Min Act: -0.0001

Data Index: 1388597 (Python Code)

Max Activating Token Index: 25

Text #18

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0259. Min Act: -0.0001

Data Index: 1135156 (C4 (Web Text))

Max Activating Token Index: 315

Text #19

Max Range: 0.0451. Min Range: -0.0451

Max Act: 0.0234. Min Act: -0.0001

Data Index: 472456 (C4 (Web Text))

Max Activating Token Index: 3