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

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

Neuron 4401 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.8159. Min Act: -0.0001

Data Index: 226779 (C4 (Web Text))

Max Activating Token Index: 515

Text #1

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.8662. Min Act: -0.0001

Data Index: 310285 (C4 (Web Text))

Max Activating Token Index: 204

Text #2

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7615. Min Act: -0.0001

Data Index: 1629547 (Python Code)

Max Activating Token Index: 860

Text #3

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.8202. Min Act: -0.0001

Data Index: 685431 (C4 (Web Text))

Max Activating Token Index: 1006

Text #4

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.8159. Min Act: -0.0001

Data Index: 1567867 (Python Code)

Max Activating Token Index: 673

Text #5

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7055. Min Act: -0.0001

Data Index: 455048 (C4 (Web Text))

Max Activating Token Index: 263

Text #6

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7340. Min Act: -0.0001

Data Index: 467584 (C4 (Web Text))

Max Activating Token Index: 302

Text #7

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7694. Min Act: -0.0001

Data Index: 1385467 (Python Code)

Max Activating Token Index: 674

Text #8

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7653. Min Act: -0.0001

Data Index: 895943 (C4 (Web Text))

Max Activating Token Index: 759

Text #9

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6895. Min Act: -0.0001

Data Index: 1234122 (C4 (Web Text))

Max Activating Token Index: 381

Text #10

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7270. Min Act: -0.0001

Data Index: 182018 (C4 (Web Text))

Max Activating Token Index: 621

Text #11

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7543. Min Act: -0.0001

Data Index: 983127 (C4 (Web Text))

Max Activating Token Index: 451

Text #12

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7392. Min Act: -0.0001

Data Index: 88485 (C4 (Web Text))

Max Activating Token Index: 768

Text #13

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.7046. Min Act: -0.0001

Data Index: 740449 (C4 (Web Text))

Max Activating Token Index: 421

Text #14

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6231. Min Act: -0.0001

Data Index: 817866 (C4 (Web Text))

Max Activating Token Index: 311

Text #15

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6657. Min Act: -0.0001

Data Index: 851247 (C4 (Web Text))

Max Activating Token Index: 573

Text #16

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.5769. Min Act: -0.0001

Data Index: 1538785 (Python Code)

Max Activating Token Index: 401

Text #17

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6718. Min Act: -0.0001

Data Index: 627172 (C4 (Web Text))

Max Activating Token Index: 812

Text #18

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6924. Min Act: -0.0001

Data Index: 612453 (C4 (Web Text))

Max Activating Token Index: 573

Text #19

Max Range: 0.8662. Min Range: -0.8662

Max Act: 0.6862. Min Act: -0.0001

Data Index: 230810 (C4 (Web Text))

Max Activating Token Index: 823