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

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

Neuron 3838 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 2.2580. Min Range: -2.2580

Max Act: 2.2580. Min Act: -0.0000

Data Index: 596634 (C4 (Web Text))

Max Activating Token Index: 868

Text #1

Max Range: 2.2580. Min Range: -2.2580

Max Act: 2.1204. Min Act: -0.0000

Data Index: 559918 (C4 (Web Text))

Max Activating Token Index: 706

Text #2

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.7005. Min Act: -0.0000

Data Index: 1373514 (Python Code)

Max Activating Token Index: 669

Text #3

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.8704. Min Act: -0.0000

Data Index: 1255409 (C4 (Web Text))

Max Activating Token Index: 388

Text #4

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.7523. Min Act: -0.0000

Data Index: 1154325 (C4 (Web Text))

Max Activating Token Index: 20

Text #5

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.8471. Min Act: -0.0000

Data Index: 1439958 (Python Code)

Max Activating Token Index: 842

Text #6

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.8113. Min Act: -0.0000

Data Index: 872458 (C4 (Web Text))

Max Activating Token Index: 872

Text #7

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.8557. Min Act: -0.0000

Data Index: 4912 (C4 (Web Text))

Max Activating Token Index: 324

Text #8

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.7513. Min Act: -0.0000

Data Index: 797205 (C4 (Web Text))

Max Activating Token Index: 16

Text #9

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.7583. Min Act: -0.0000

Data Index: 327491 (C4 (Web Text))

Max Activating Token Index: 655

Text #10

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.7737. Min Act: -0.0000

Data Index: 1475395 (Python Code)

Max Activating Token Index: 706

Text #11

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.5906. Min Act: -0.0000

Data Index: 1449960 (Python Code)

Max Activating Token Index: 348

Text #12

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6747. Min Act: -0.0000

Data Index: 1112653 (C4 (Web Text))

Max Activating Token Index: 668

Text #13

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6631. Min Act: -0.0000

Data Index: 1004132 (C4 (Web Text))

Max Activating Token Index: 11

Text #14

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6460. Min Act: -0.0000

Data Index: 344315 (C4 (Web Text))

Max Activating Token Index: 641

Text #15

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.4526. Min Act: -0.0000

Data Index: 1415414 (Python Code)

Max Activating Token Index: 291

Text #16

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6332. Min Act: -0.0000

Data Index: 98403 (C4 (Web Text))

Max Activating Token Index: 704

Text #17

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6105. Min Act: -0.0000

Data Index: 814438 (C4 (Web Text))

Max Activating Token Index: 921

Text #18

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.5394. Min Act: -0.0000

Data Index: 1075493 (C4 (Web Text))

Max Activating Token Index: 286

Text #19

Max Range: 2.2580. Min Range: -2.2580

Max Act: 1.6185. Min Act: -0.0000

Data Index: 1230666 (C4 (Web Text))

Max Activating Token Index: 505