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

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

Neuron 1054 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.1938. Min Range: -4.1938

Max Act: 4.1938. Min Act: -0.0000

Data Index: 89308 (C4 (Web Text))

Max Activating Token Index: 138

Text #1

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.7393. Min Act: -0.0000

Data Index: 940977 (C4 (Web Text))

Max Activating Token Index: 836

Text #2

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.8494. Min Act: -0.0000

Data Index: 1348552 (C4 (Web Text))

Max Activating Token Index: 653

Text #3

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.3973. Min Act: -0.0000

Data Index: 44342 (C4 (Web Text))

Max Activating Token Index: 680

Text #4

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.7602. Min Act: -0.0000

Data Index: 797284 (C4 (Web Text))

Max Activating Token Index: 1022

Text #5

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.3852. Min Act: -0.0000

Data Index: 1273765 (C4 (Web Text))

Max Activating Token Index: 679

Text #6

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.5081. Min Act: -0.0000

Data Index: 129802 (C4 (Web Text))

Max Activating Token Index: 210

Text #7

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.4827. Min Act: -0.0000

Data Index: 32052 (C4 (Web Text))

Max Activating Token Index: 572

Text #8

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.2028. Min Act: -0.0000

Data Index: 1595193 (Python Code)

Max Activating Token Index: 1013

Text #9

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.4067. Min Act: -0.0000

Data Index: 487206 (C4 (Web Text))

Max Activating Token Index: 504

Text #10

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.3656. Min Act: -0.0000

Data Index: 345216 (C4 (Web Text))

Max Activating Token Index: 895

Text #11

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.3253. Min Act: -0.0000

Data Index: 564558 (C4 (Web Text))

Max Activating Token Index: 259

Text #12

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.3900. Min Act: -0.0000

Data Index: 1274215 (C4 (Web Text))

Max Activating Token Index: 530

Text #13

Max Range: 4.1938. Min Range: -4.1938

Max Act: 2.7285. Min Act: -0.0000

Data Index: 805170 (C4 (Web Text))

Max Activating Token Index: 786

Text #14

Max Range: 4.1938. Min Range: -4.1938

Max Act: 2.9575. Min Act: -0.0000

Data Index: 191658 (C4 (Web Text))

Max Activating Token Index: 656

Text #15

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.0794. Min Act: -0.0000

Data Index: 1478475 (Python Code)

Max Activating Token Index: 924

Text #16

Max Range: 4.1938. Min Range: -4.1938

Max Act: 3.0066. Min Act: -0.0000

Data Index: 709872 (C4 (Web Text))

Max Activating Token Index: 759

Text #17

Max Range: 4.1938. Min Range: -4.1938

Max Act: 2.8538. Min Act: -0.0000

Data Index: 782391 (C4 (Web Text))

Max Activating Token Index: 642

Text #18

Max Range: 4.1938. Min Range: -4.1938

Max Act: 2.8278. Min Act: -0.0000

Data Index: 1595939 (Python Code)

Max Activating Token Index: 773

Text #19

Max Range: 4.1938. Min Range: -4.1938

Max Act: 2.9723. Min Act: -0.0000

Data Index: 570070 (C4 (Web Text))

Max Activating Token Index: 817