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

Dataset: The Pile

Neuron 3219 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.5017. Min Range: -3.5017

Max Act: 3.4617. Min Act: -0.0000

Data Index: 1502259 (The Pile)

Max Activating Token Index: 5

Text #1

Max Range: 3.5017. Min Range: -3.5017

Max Act: 3.5017. Min Act: -0.0000

Data Index: 276820 (The Pile)

Max Activating Token Index: 562

Text #2

Max Range: 3.5017. Min Range: -3.5017

Max Act: 3.2964. Min Act: -0.0000

Data Index: 670147 (The Pile)

Max Activating Token Index: 807

Text #3

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.9014. Min Act: -0.0000

Data Index: 1603303 (The Pile)

Max Activating Token Index: 779

Text #4

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.5668. Min Act: -0.0000

Data Index: 1074515 (The Pile)

Max Activating Token Index: 1003

Text #5

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6969. Min Act: -0.0000

Data Index: 1523304 (The Pile)

Max Activating Token Index: 66

Text #6

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.9653. Min Act: -0.0000

Data Index: 494277 (The Pile)

Max Activating Token Index: 691

Text #7

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.7581. Min Act: -0.0000

Data Index: 1423022 (The Pile)

Max Activating Token Index: 980

Text #8

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6271. Min Act: -0.0000

Data Index: 1000544 (The Pile)

Max Activating Token Index: 337

Text #9

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.8807. Min Act: -0.0000

Data Index: 1288674 (The Pile)

Max Activating Token Index: 203

Text #10

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.5860. Min Act: -0.0000

Data Index: 1651108 (The Pile)

Max Activating Token Index: 453

Text #11

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.7121. Min Act: -0.0000

Data Index: 1124967 (The Pile)

Max Activating Token Index: 2

Text #12

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.5119. Min Act: -0.0000

Data Index: 1291860 (The Pile)

Max Activating Token Index: 980

Text #13

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.8271. Min Act: -0.0000

Data Index: 1605027 (The Pile)

Max Activating Token Index: 550

Text #14

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6840. Min Act: -0.0000

Data Index: 1534502 (The Pile)

Max Activating Token Index: 495

Text #15

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6593. Min Act: -0.0000

Data Index: 39603 (The Pile)

Max Activating Token Index: 913

Text #16

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.7738. Min Act: -0.0000

Data Index: 1832982 (The Pile)

Max Activating Token Index: 130

Text #17

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6309. Min Act: -0.0000

Data Index: 1998796 (The Pile)

Max Activating Token Index: 434

Text #18

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.6176. Min Act: -0.0000

Data Index: 1256107 (The Pile)

Max Activating Token Index: 379

Text #19

Max Range: 3.5017. Min Range: -3.5017

Max Act: 2.5846. Min Act: -0.0000

Data Index: 848718 (The Pile)

Max Activating Token Index: 400