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

Dataset: The Pile

Neuron 969 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 14.1750. Min Range: -14.1750

Max Act: 14.1750. Min Act: -0.0000

Data Index: 841965 (The Pile)

Max Activating Token Index: 492

Text #1

Max Range: 14.1750. Min Range: -14.1750

Max Act: 13.6658. Min Act: -0.0000

Data Index: 762437 (The Pile)

Max Activating Token Index: 440

Text #2

Max Range: 14.1750. Min Range: -14.1750

Max Act: 9.2717. Min Act: -0.0000

Data Index: 8112 (The Pile)

Max Activating Token Index: 233

Text #3

Max Range: 14.1750. Min Range: -14.1750

Max Act: 9.3610. Min Act: -0.0000

Data Index: 968730 (The Pile)

Max Activating Token Index: 519

Text #4

Max Range: 14.1750. Min Range: -14.1750

Max Act: 7.6454. Min Act: -0.0000

Data Index: 205118 (The Pile)

Max Activating Token Index: 444

Text #5

Max Range: 14.1750. Min Range: -14.1750

Max Act: 4.3639. Min Act: -0.0000

Data Index: 22375 (The Pile)

Max Activating Token Index: 525

Text #6

Max Range: 14.1750. Min Range: -14.1750

Max Act: 6.9040. Min Act: -0.0000

Data Index: 1356379 (The Pile)

Max Activating Token Index: 333

Text #7

Max Range: 14.1750. Min Range: -14.1750

Max Act: 1.1425. Min Act: -0.0000

Data Index: 695179 (The Pile)

Max Activating Token Index: 504

Text #8

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.1732. Min Act: -0.0000

Data Index: 66771 (The Pile)

Max Activating Token Index: 357

Text #9

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.4597. Min Act: -0.0000

Data Index: 741577 (The Pile)

Max Activating Token Index: 162

Text #10

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.8990. Min Act: -0.0000

Data Index: 174910 (The Pile)

Max Activating Token Index: 595

Text #11

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.0901. Min Act: -0.0000

Data Index: 1485743 (The Pile)

Max Activating Token Index: 506

Text #12

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.2436. Min Act: -0.0000

Data Index: 1939417 (The Pile)

Max Activating Token Index: 125

Text #13

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.0808. Min Act: -0.0000

Data Index: 1817031 (The Pile)

Max Activating Token Index: 312

Text #14

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.2144. Min Act: -0.0000

Data Index: 1053821 (The Pile)

Max Activating Token Index: 158

Text #15

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.0343. Min Act: -0.0000

Data Index: 1678868 (The Pile)

Max Activating Token Index: 432

Text #16

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.2115. Min Act: -0.0000

Data Index: 1206801 (The Pile)

Max Activating Token Index: 67

Text #17

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.1472. Min Act: -0.0000

Data Index: 397587 (The Pile)

Max Activating Token Index: 468

Text #18

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.1567. Min Act: -0.0000

Data Index: 1239723 (The Pile)

Max Activating Token Index: 895

Text #19

Max Range: 14.1750. Min Range: -14.1750

Max Act: 0.3057. Min Act: -0.0000

Data Index: 42437 (The Pile)

Max Activating Token Index: 103