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

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

Neuron 2450 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0371. Min Act: -0.0002

Data Index: 1118084 (C4 (Web Text))

Max Activating Token Index: 133

Text #1

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0410. Min Act: -0.0002

Data Index: 362354 (C4 (Web Text))

Max Activating Token Index: 290

Text #2

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0358. Min Act: -0.0002

Data Index: 495553 (C4 (Web Text))

Max Activating Token Index: 189

Text #3

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0337. Min Act: -0.0002

Data Index: 144255 (C4 (Web Text))

Max Activating Token Index: 125

Text #4

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0326. Min Act: -0.0002

Data Index: 725198 (C4 (Web Text))

Max Activating Token Index: 332

Text #5

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0324. Min Act: -0.0002

Data Index: 163716 (C4 (Web Text))

Max Activating Token Index: 357

Text #6

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0309. Min Act: -0.0002

Data Index: 282043 (C4 (Web Text))

Max Activating Token Index: 286

Text #7

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0325. Min Act: -0.0002

Data Index: 331370 (C4 (Web Text))

Max Activating Token Index: 251

Text #8

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0316. Min Act: -0.0002

Data Index: 1313061 (C4 (Web Text))

Max Activating Token Index: 434

Text #9

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0272. Min Act: -0.0002

Data Index: 857713 (C4 (Web Text))

Max Activating Token Index: 493

Text #10

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0259. Min Act: -0.0002

Data Index: 1301883 (C4 (Web Text))

Max Activating Token Index: 640

Text #11

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0284. Min Act: -0.0002

Data Index: 1444758 (Python Code)

Max Activating Token Index: 358

Text #12

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0292. Min Act: -0.0002

Data Index: 457754 (C4 (Web Text))

Max Activating Token Index: 604

Text #13

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0260. Min Act: -0.0002

Data Index: 1433336 (Python Code)

Max Activating Token Index: 504

Text #14

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0253. Min Act: -0.0002

Data Index: 914528 (C4 (Web Text))

Max Activating Token Index: 590

Text #15

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0228. Min Act: -0.0002

Data Index: 1079081 (C4 (Web Text))

Max Activating Token Index: 652

Text #16

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0233. Min Act: -0.0002

Data Index: 446363 (C4 (Web Text))

Max Activating Token Index: 839

Text #17

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0232. Min Act: -0.0002

Data Index: 92864 (C4 (Web Text))

Max Activating Token Index: 909

Text #18

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0232. Min Act: -0.0002

Data Index: 892918 (C4 (Web Text))

Max Activating Token Index: 524

Text #19

Max Range: 0.0410. Min Range: -0.0410

Max Act: 0.0236. Min Act: -0.0002

Data Index: 1499093 (Python Code)

Max Activating Token Index: 672