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

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

Neuron 2545 in Layer 0

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0670. Min Act: -0.0001

Data Index: 1497021 (Python Code)

Max Activating Token Index: 23

Text #1

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0680. Min Act: -0.0001

Data Index: 1387859 (Python Code)

Max Activating Token Index: 33

Text #2

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0652. Min Act: -0.0002

Data Index: 1457918 (Python Code)

Max Activating Token Index: 25

Text #3

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0650. Min Act: -0.0002

Data Index: 757168 (C4 (Web Text))

Max Activating Token Index: 144

Text #4

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0638. Min Act: -0.0002

Data Index: 1194852 (C4 (Web Text))

Max Activating Token Index: 960

Text #5

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0620. Min Act: -0.0001

Data Index: 1465164 (Python Code)

Max Activating Token Index: 20

Text #6

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0619. Min Act: -0.0001

Data Index: 1387473 (Python Code)

Max Activating Token Index: 51

Text #7

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0614. Min Act: -0.0002

Data Index: 1339084 (C4 (Web Text))

Max Activating Token Index: 151

Text #8

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0614. Min Act: -0.0001

Data Index: 1370480 (Python Code)

Max Activating Token Index: 17

Text #9

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0613. Min Act: -0.0002

Data Index: 1359043 (C4 (Web Text))

Max Activating Token Index: 121

Text #10

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0588. Min Act: -0.0001

Data Index: 1635293 (Python Code)

Max Activating Token Index: 20

Text #11

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0610. Min Act: -0.0002

Data Index: 1078686 (C4 (Web Text))

Max Activating Token Index: 255

Text #12

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0605. Min Act: -0.0001

Data Index: 1512525 (Python Code)

Max Activating Token Index: 29

Text #13

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0604. Min Act: -0.0001

Data Index: 1611452 (Python Code)

Max Activating Token Index: 33

Text #14

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0603. Min Act: -0.0001

Data Index: 1639553 (Python Code)

Max Activating Token Index: 36

Text #15

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0601. Min Act: -0.0002

Data Index: 777974 (C4 (Web Text))

Max Activating Token Index: 160

Text #16

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0602. Min Act: -0.0002

Data Index: 1129872 (C4 (Web Text))

Max Activating Token Index: 644

Text #17

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0576. Min Act: -0.0001

Data Index: 1595014 (Python Code)

Max Activating Token Index: 36

Text #18

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0593. Min Act: -0.0002

Data Index: 1440901 (Python Code)

Max Activating Token Index: 29

Text #19

Max Range: 0.0680. Min Range: -0.0680

Max Act: 0.0593. Min Act: -0.0001

Data Index: 1472352 (Python Code)

Max Activating Token Index: 26