Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 53 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0503. Min Act: -0.0002

Data Index: 837055 (C4 (Web Text))

Max Activating Token Index: 108

Text #1

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0486. Min Act: -0.0002

Data Index: 732898 (C4 (Web Text))

Max Activating Token Index: 826

Text #2

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0480. Min Act: -0.0002

Data Index: 799886 (C4 (Web Text))

Max Activating Token Index: 844

Text #3

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0503. Min Act: -0.0002

Data Index: 9939 (C4 (Web Text))

Max Activating Token Index: 967

Text #4

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0486. Min Act: -0.0002

Data Index: 779287 (C4 (Web Text))

Max Activating Token Index: 788

Text #5

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0474. Min Act: -0.0002

Data Index: 588107 (C4 (Web Text))

Max Activating Token Index: 268

Text #6

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0469. Min Act: -0.0001

Data Index: 1046645 (C4 (Web Text))

Max Activating Token Index: 622

Text #7

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0485. Min Act: -0.0002

Data Index: 1276838 (C4 (Web Text))

Max Activating Token Index: 724

Text #8

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0483. Min Act: -0.0001

Data Index: 928496 (C4 (Web Text))

Max Activating Token Index: 759

Text #9

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0477. Min Act: -0.0002

Data Index: 785548 (C4 (Web Text))

Max Activating Token Index: 287

Text #10

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0469. Min Act: -0.0002

Data Index: 1017175 (C4 (Web Text))

Max Activating Token Index: 939

Text #11

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0478. Min Act: -0.0002

Data Index: 612997 (C4 (Web Text))

Max Activating Token Index: 978

Text #12

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0470. Min Act: -0.0002

Data Index: 1299826 (C4 (Web Text))

Max Activating Token Index: 906

Text #13

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0466. Min Act: -0.0002

Data Index: 250845 (C4 (Web Text))

Max Activating Token Index: 681

Text #14

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0473. Min Act: -0.0002

Data Index: 762955 (C4 (Web Text))

Max Activating Token Index: 177

Text #15

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0470. Min Act: -0.0002

Data Index: 458390 (C4 (Web Text))

Max Activating Token Index: 992

Text #16

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0486. Min Act: -0.0002

Data Index: 685104 (C4 (Web Text))

Max Activating Token Index: 205

Text #17

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0472. Min Act: -0.0001

Data Index: 584844 (C4 (Web Text))

Max Activating Token Index: 414

Text #18

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0468. Min Act: -0.0002

Data Index: 591 (C4 (Web Text))

Max Activating Token Index: 924

Text #19

Max Range: 0.0503. Min Range: -0.0503

Max Act: 0.0468. Min Act: -0.0002

Data Index: 625764 (C4 (Web Text))

Max Activating Token Index: 300