Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3560 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0750. Min Act: -0.0001

Data Index: 794584 (C4 (Web Text))

Max Activating Token Index: 64

Text #1

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0698. Min Act: -0.0001

Data Index: 601610 (C4 (Web Text))

Max Activating Token Index: 105

Text #2

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0655. Min Act: -0.0001

Data Index: 1305114 (C4 (Web Text))

Max Activating Token Index: 857

Text #3

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0645. Min Act: -0.0002

Data Index: 1108908 (C4 (Web Text))

Max Activating Token Index: 692

Text #4

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0666. Min Act: -0.0001

Data Index: 1445448 (Python Code)

Max Activating Token Index: 223

Text #5

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0645. Min Act: -0.0001

Data Index: 1082437 (C4 (Web Text))

Max Activating Token Index: 487

Text #6

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0623. Min Act: -0.0002

Data Index: 274506 (C4 (Web Text))

Max Activating Token Index: 672

Text #7

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0592. Min Act: -0.0001

Data Index: 123930 (C4 (Web Text))

Max Activating Token Index: 968

Text #8

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0548. Min Act: -0.0001

Data Index: 1649093 (Python Code)

Max Activating Token Index: 164

Text #9

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0541. Min Act: -0.0001

Data Index: 850302 (C4 (Web Text))

Max Activating Token Index: 744

Text #10

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0585. Min Act: -0.0001

Data Index: 1144053 (C4 (Web Text))

Max Activating Token Index: 805

Text #11

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0602. Min Act: -0.0001

Data Index: 624391 (C4 (Web Text))

Max Activating Token Index: 20

Text #12

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0574. Min Act: -0.0001

Data Index: 1512047 (Python Code)

Max Activating Token Index: 821

Text #13

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0596. Min Act: -0.0001

Data Index: 1639819 (Python Code)

Max Activating Token Index: 61

Text #14

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0599. Min Act: -0.0001

Data Index: 987690 (C4 (Web Text))

Max Activating Token Index: 273

Text #15

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0640. Min Act: -0.0001

Data Index: 1384846 (Python Code)

Max Activating Token Index: 324

Text #16

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0565. Min Act: -0.0001

Data Index: 524091 (C4 (Web Text))

Max Activating Token Index: 130

Text #17

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0575. Min Act: -0.0001

Data Index: 1566648 (Python Code)

Max Activating Token Index: 626

Text #18

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0604. Min Act: -0.0001

Data Index: 1157355 (C4 (Web Text))

Max Activating Token Index: 105

Text #19

Max Range: 0.0750. Min Range: -0.0750

Max Act: 0.0581. Min Act: -0.0001

Data Index: 1042935 (C4 (Web Text))

Max Activating Token Index: 759