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

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

Neuron 4747 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0214. Min Range: -1.0214

Max Act: 1.0214. Min Act: -0.0002

Data Index: 1193876 (C4 (Web Text))

Max Activating Token Index: 363

Text #1

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7689. Min Act: -0.0002

Data Index: 561749 (C4 (Web Text))

Max Activating Token Index: 319

Text #2

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8167. Min Act: -0.0002

Data Index: 455006 (C4 (Web Text))

Max Activating Token Index: 664

Text #3

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8337. Min Act: -0.0002

Data Index: 71988 (C4 (Web Text))

Max Activating Token Index: 309

Text #4

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8739. Min Act: -0.0002

Data Index: 508077 (C4 (Web Text))

Max Activating Token Index: 316

Text #5

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8724. Min Act: -0.0002

Data Index: 1190919 (C4 (Web Text))

Max Activating Token Index: 422

Text #6

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7457. Min Act: -0.0002

Data Index: 41327 (C4 (Web Text))

Max Activating Token Index: 286

Text #7

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7310. Min Act: -0.0002

Data Index: 1322016 (C4 (Web Text))

Max Activating Token Index: 376

Text #8

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8642. Min Act: -0.0002

Data Index: 665190 (C4 (Web Text))

Max Activating Token Index: 192

Text #9

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8825. Min Act: -0.0002

Data Index: 1160000 (C4 (Web Text))

Max Activating Token Index: 685

Text #10

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8153. Min Act: -0.0002

Data Index: 940028 (C4 (Web Text))

Max Activating Token Index: 636

Text #11

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7409. Min Act: -0.0002

Data Index: 1359276 (C4 (Web Text))

Max Activating Token Index: 60

Text #12

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.8813. Min Act: -0.0002

Data Index: 562610 (C4 (Web Text))

Max Activating Token Index: 812

Text #13

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.6409. Min Act: -0.0002

Data Index: 204487 (C4 (Web Text))

Max Activating Token Index: 387

Text #14

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.6565. Min Act: -0.0002

Data Index: 498597 (C4 (Web Text))

Max Activating Token Index: 417

Text #15

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7272. Min Act: -0.0002

Data Index: 1596874 (Python Code)

Max Activating Token Index: 338

Text #16

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.6943. Min Act: -0.0002

Data Index: 935584 (C4 (Web Text))

Max Activating Token Index: 122

Text #17

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.6559. Min Act: -0.0002

Data Index: 171503 (C4 (Web Text))

Max Activating Token Index: 580

Text #18

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7339. Min Act: -0.0002

Data Index: 1492295 (Python Code)

Max Activating Token Index: 850

Text #19

Max Range: 1.0214. Min Range: -1.0214

Max Act: 0.7331. Min Act: -0.0002

Data Index: 635602 (C4 (Web Text))

Max Activating Token Index: 971