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

Dataset: The Pile

Neuron 3904 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.7266. Min Act: -0.0000

Data Index: 1509963 (The Pile)

Max Activating Token Index: 552

Text #1

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6820. Min Act: -0.0000

Data Index: 71051 (The Pile)

Max Activating Token Index: 37

Text #2

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6422. Min Act: -0.0000

Data Index: 326667 (The Pile)

Max Activating Token Index: 760

Text #3

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6514. Min Act: -0.0000

Data Index: 1555371 (The Pile)

Max Activating Token Index: 138

Text #4

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6229. Min Act: -0.0000

Data Index: 555849 (The Pile)

Max Activating Token Index: 899

Text #5

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6229. Min Act: -0.0000

Data Index: 769845 (The Pile)

Max Activating Token Index: 1000

Text #6

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6149. Min Act: -0.0000

Data Index: 1643278 (The Pile)

Max Activating Token Index: 862

Text #7

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6128. Min Act: -0.0000

Data Index: 72764 (The Pile)

Max Activating Token Index: 753

Text #8

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6073. Min Act: -0.0000

Data Index: 1782055 (The Pile)

Max Activating Token Index: 281

Text #9

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5766. Min Act: -0.0000

Data Index: 309674 (The Pile)

Max Activating Token Index: 589

Text #10

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.6094. Min Act: -0.0000

Data Index: 1433444 (The Pile)

Max Activating Token Index: 524

Text #11

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5571. Min Act: 0.0000

Data Index: 1274261 (The Pile)

Max Activating Token Index: 25

Text #12

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5510. Min Act: -0.0000

Data Index: 1283481 (The Pile)

Max Activating Token Index: 206

Text #13

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5654. Min Act: -0.0000

Data Index: 573235 (The Pile)

Max Activating Token Index: 657

Text #14

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5732. Min Act: -0.0000

Data Index: 1523411 (The Pile)

Max Activating Token Index: 877

Text #15

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5541. Min Act: -0.0000

Data Index: 85714 (The Pile)

Max Activating Token Index: 278

Text #16

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5698. Min Act: -0.0000

Data Index: 560351 (The Pile)

Max Activating Token Index: 498

Text #17

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5697. Min Act: -0.0000

Data Index: 1240084 (The Pile)

Max Activating Token Index: 52

Text #18

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5298. Min Act: -0.0000

Data Index: 1066890 (The Pile)

Max Activating Token Index: 795

Text #19

Max Range: 0.7266. Min Range: -0.7266

Max Act: 0.5412. Min Act: -0.0000

Data Index: 1412417 (The Pile)

Max Activating Token Index: 483