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

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

Neuron 3982 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8339. Min Range: -2.8339

Max Act: 2.8339. Min Act: -0.0000

Data Index: 756460 (C4 (Web Text))

Max Activating Token Index: 619

Text #1

Max Range: 2.8339. Min Range: -2.8339

Max Act: 2.4270. Min Act: -0.0000

Data Index: 1293816 (C4 (Web Text))

Max Activating Token Index: 249

Text #2

Max Range: 2.8339. Min Range: -2.8339

Max Act: 2.0296. Min Act: -0.0000

Data Index: 227428 (C4 (Web Text))

Max Activating Token Index: 395

Text #3

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.9608. Min Act: -0.0000

Data Index: 275862 (C4 (Web Text))

Max Activating Token Index: 322

Text #4

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.7631. Min Act: -0.0000

Data Index: 546557 (C4 (Web Text))

Max Activating Token Index: 891

Text #5

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.6418. Min Act: -0.0000

Data Index: 1046369 (C4 (Web Text))

Max Activating Token Index: 456

Text #6

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4828. Min Act: -0.0001

Data Index: 138694 (C4 (Web Text))

Max Activating Token Index: 875

Text #7

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4732. Min Act: -0.0000

Data Index: 287348 (C4 (Web Text))

Max Activating Token Index: 351

Text #8

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4648. Min Act: -0.0000

Data Index: 1623497 (Python Code)

Max Activating Token Index: 924

Text #9

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4324. Min Act: -0.0000

Data Index: 38249 (C4 (Web Text))

Max Activating Token Index: 147

Text #10

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4019. Min Act: -0.0000

Data Index: 1582909 (Python Code)

Max Activating Token Index: 693

Text #11

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.4105. Min Act: -0.0000

Data Index: 118124 (C4 (Web Text))

Max Activating Token Index: 451

Text #12

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.3335. Min Act: -0.0000

Data Index: 794276 (C4 (Web Text))

Max Activating Token Index: 87

Text #13

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.3431. Min Act: -0.0000

Data Index: 1071095 (C4 (Web Text))

Max Activating Token Index: 213

Text #14

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.2158. Min Act: -0.0000

Data Index: 898583 (C4 (Web Text))

Max Activating Token Index: 827

Text #15

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.2340. Min Act: -0.0000

Data Index: 1442547 (Python Code)

Max Activating Token Index: 331

Text #16

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.3493. Min Act: -0.0000

Data Index: 1286121 (C4 (Web Text))

Max Activating Token Index: 753

Text #17

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.2722. Min Act: -0.0000

Data Index: 950255 (C4 (Web Text))

Max Activating Token Index: 508

Text #18

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.2623. Min Act: -0.0000

Data Index: 1250438 (C4 (Web Text))

Max Activating Token Index: 957

Text #19

Max Range: 2.8339. Min Range: -2.8339

Max Act: 1.2178. Min Act: -0.0000

Data Index: 1166290 (C4 (Web Text))

Max Activating Token Index: 780