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

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

Neuron 5065 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.6639. Min Act: 0.0000

Data Index: 944532 (C4 (Web Text))

Max Activating Token Index: 397

Text #1

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.6223. Min Act: -0.0000

Data Index: 382670 (C4 (Web Text))

Max Activating Token Index: 505

Text #2

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.8297. Min Act: 0.0000

Data Index: 501230 (C4 (Web Text))

Max Activating Token Index: 304

Text #3

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.3748. Min Act: -0.0000

Data Index: 676288 (C4 (Web Text))

Max Activating Token Index: 520

Text #4

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0821. Min Act: -0.0000

Data Index: 1324918 (C4 (Web Text))

Max Activating Token Index: 405

Text #5

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0889. Min Act: -0.0000

Data Index: 549733 (C4 (Web Text))

Max Activating Token Index: 846

Text #6

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0795. Min Act: -0.0000

Data Index: 774509 (C4 (Web Text))

Max Activating Token Index: 701

Text #7

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0542. Min Act: -0.0000

Data Index: 898284 (C4 (Web Text))

Max Activating Token Index: 475

Text #8

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.2809. Min Act: -0.0000

Data Index: 844927 (C4 (Web Text))

Max Activating Token Index: 323

Text #9

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.1250. Min Act: 0.0000

Data Index: 1011982 (C4 (Web Text))

Max Activating Token Index: 476

Text #10

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.1176. Min Act: -0.0000

Data Index: 1067480 (C4 (Web Text))

Max Activating Token Index: 348

Text #11

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0783. Min Act: 0.0001

Data Index: 946403 (C4 (Web Text))

Max Activating Token Index: 477

Text #12

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.1612. Min Act: 0.0000

Data Index: 633856 (C4 (Web Text))

Max Activating Token Index: 7

Text #13

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0324. Min Act: -0.0000

Data Index: 721321 (C4 (Web Text))

Max Activating Token Index: 816

Text #14

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0429. Min Act: -0.0000

Data Index: 1324072 (C4 (Web Text))

Max Activating Token Index: 605

Text #15

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0239. Min Act: -0.0000

Data Index: 1504240 (Python Code)

Max Activating Token Index: 799

Text #16

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0970. Min Act: -0.0000

Data Index: 57310 (C4 (Web Text))

Max Activating Token Index: 46

Text #17

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.1141. Min Act: 0.0000

Data Index: 911323 (C4 (Web Text))

Max Activating Token Index: 449

Text #18

Max Range: 2.8297. Min Range: -2.8297

Max Act: 2.0038. Min Act: -0.0000

Data Index: 1479400 (Python Code)

Max Activating Token Index: 601

Text #19

Max Range: 2.8297. Min Range: -2.8297

Max Act: 1.9907. Min Act: 0.0001

Data Index: 154255 (C4 (Web Text))

Max Activating Token Index: 547