Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 2006 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1972. Min Act: -0.0001

Data Index: 1356073 (C4 (Web Text))

Max Activating Token Index: 509

Text #1

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1797. Min Act: -0.0001

Data Index: 1196519 (C4 (Web Text))

Max Activating Token Index: 462

Text #2

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1745. Min Act: -0.0001

Data Index: 152899 (C4 (Web Text))

Max Activating Token Index: 454

Text #3

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1668. Min Act: -0.0001

Data Index: 736998 (C4 (Web Text))

Max Activating Token Index: 840

Text #4

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1667. Min Act: -0.0001

Data Index: 53331 (C4 (Web Text))

Max Activating Token Index: 997

Text #5

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1598. Min Act: -0.0001

Data Index: 690038 (C4 (Web Text))

Max Activating Token Index: 584

Text #6

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1648. Min Act: -0.0001

Data Index: 461277 (C4 (Web Text))

Max Activating Token Index: 164

Text #7

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1585. Min Act: -0.0001

Data Index: 1101562 (C4 (Web Text))

Max Activating Token Index: 120

Text #8

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1553. Min Act: -0.0001

Data Index: 464864 (C4 (Web Text))

Max Activating Token Index: 257

Text #9

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1611. Min Act: -0.0001

Data Index: 1204399 (C4 (Web Text))

Max Activating Token Index: 792

Text #10

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1529. Min Act: -0.0001

Data Index: 30666 (C4 (Web Text))

Max Activating Token Index: 353

Text #11

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1581. Min Act: -0.0001

Data Index: 456326 (C4 (Web Text))

Max Activating Token Index: 928

Text #12

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1539. Min Act: -0.0001

Data Index: 864194 (C4 (Web Text))

Max Activating Token Index: 274

Text #13

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1504. Min Act: -0.0001

Data Index: 1258010 (C4 (Web Text))

Max Activating Token Index: 994

Text #14

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1608. Min Act: -0.0001

Data Index: 1319558 (C4 (Web Text))

Max Activating Token Index: 592

Text #15

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1632. Min Act: -0.0001

Data Index: 113746 (C4 (Web Text))

Max Activating Token Index: 854

Text #16

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1505. Min Act: -0.0001

Data Index: 827711 (C4 (Web Text))

Max Activating Token Index: 882

Text #17

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1508. Min Act: -0.0001

Data Index: 1058302 (C4 (Web Text))

Max Activating Token Index: 14

Text #18

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1400. Min Act: -0.0001

Data Index: 1109363 (C4 (Web Text))

Max Activating Token Index: 831

Text #19

Max Range: 0.1972. Min Range: -0.1972

Max Act: 0.1539. Min Act: -0.0001

Data Index: 820807 (C4 (Web Text))

Max Activating Token Index: 340