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

Dataset: The Pile

Neuron 4851 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.2857. Min Act: -0.0000

Data Index: 790984 (The Pile)

Max Activating Token Index: 345

Text #1

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.4769. Min Act: -0.0000

Data Index: 1700370 (The Pile)

Max Activating Token Index: 858

Text #2

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.7922. Min Act: -0.0000

Data Index: 1842966 (The Pile)

Max Activating Token Index: 148

Text #3

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.1001. Min Act: -0.0000

Data Index: 582138 (The Pile)

Max Activating Token Index: 434

Text #4

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.1528. Min Act: -0.0000

Data Index: 868343 (The Pile)

Max Activating Token Index: 846

Text #5

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.9134. Min Act: -0.0000

Data Index: 839623 (The Pile)

Max Activating Token Index: 812

Text #6

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.2337. Min Act: -0.0000

Data Index: 507462 (The Pile)

Max Activating Token Index: 604

Text #7

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.1191. Min Act: -0.0000

Data Index: 552754 (The Pile)

Max Activating Token Index: 673

Text #8

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.9384. Min Act: -0.0000

Data Index: 1693988 (The Pile)

Max Activating Token Index: 478

Text #9

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.0060. Min Act: -0.0000

Data Index: 992257 (The Pile)

Max Activating Token Index: 810

Text #10

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.7539. Min Act: -0.0000

Data Index: 1168792 (The Pile)

Max Activating Token Index: 644

Text #11

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.0796. Min Act: -0.0000

Data Index: 1354213 (The Pile)

Max Activating Token Index: 496

Text #12

Max Range: 5.7922. Min Range: -5.7922

Max Act: 5.0204. Min Act: -0.0000

Data Index: 1669257 (The Pile)

Max Activating Token Index: 908

Text #13

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.8134. Min Act: -0.0000

Data Index: 995442 (The Pile)

Max Activating Token Index: 600

Text #14

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.4907. Min Act: -0.0000

Data Index: 136921 (The Pile)

Max Activating Token Index: 712

Text #15

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.3207. Min Act: -0.0000

Data Index: 1966824 (The Pile)

Max Activating Token Index: 563

Text #16

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.2577. Min Act: -0.0000

Data Index: 1725874 (The Pile)

Max Activating Token Index: 594

Text #17

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.0573. Min Act: -0.0000

Data Index: 845404 (The Pile)

Max Activating Token Index: 1008

Text #18

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.0363. Min Act: -0.0000

Data Index: 1112927 (The Pile)

Max Activating Token Index: 354

Text #19

Max Range: 5.7922. Min Range: -5.7922

Max Act: 4.1272. Min Act: -0.0000

Data Index: 1222749 (The Pile)

Max Activating Token Index: 320