Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: The Pile

Neuron 26 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l-pile')

Text #0

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.5256. Min Act: -0.0001

Data Index: 1542136 (The Pile)

Max Activating Token Index: 460

Text #1

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4858. Min Act: -0.0001

Data Index: 118813 (The Pile)

Max Activating Token Index: 211

Text #2

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4377. Min Act: -0.0000

Data Index: 1441975 (The Pile)

Max Activating Token Index: 922

Text #3

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4240. Min Act: -0.0000

Data Index: 1843439 (The Pile)

Max Activating Token Index: 977

Text #4

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4433. Min Act: -0.0000

Data Index: 527517 (The Pile)

Max Activating Token Index: 657

Text #5

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4395. Min Act: -0.0001

Data Index: 1510793 (The Pile)

Max Activating Token Index: 456

Text #6

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4211. Min Act: -0.0000

Data Index: 346548 (The Pile)

Max Activating Token Index: 729

Text #7

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4297. Min Act: -0.0000

Data Index: 1883105 (The Pile)

Max Activating Token Index: 656

Text #8

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4209. Min Act: -0.0000

Data Index: 1207525 (The Pile)

Max Activating Token Index: 92

Text #9

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4101. Min Act: -0.0000

Data Index: 630545 (The Pile)

Max Activating Token Index: 983

Text #10

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4219. Min Act: -0.0001

Data Index: 750136 (The Pile)

Max Activating Token Index: 737

Text #11

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4323. Min Act: -0.0000

Data Index: 749374 (The Pile)

Max Activating Token Index: 473

Text #12

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4110. Min Act: -0.0001

Data Index: 1654180 (The Pile)

Max Activating Token Index: 601

Text #13

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.3959. Min Act: -0.0000

Data Index: 1134904 (The Pile)

Max Activating Token Index: 380

Text #14

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4049. Min Act: -0.0001

Data Index: 750586 (The Pile)

Max Activating Token Index: 172

Text #15

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.4032. Min Act: -0.0000

Data Index: 1648479 (The Pile)

Max Activating Token Index: 577

Text #16

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.3967. Min Act: -0.0001

Data Index: 1340715 (The Pile)

Max Activating Token Index: 834

Text #17

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.3979. Min Act: -0.0000

Data Index: 1620287 (The Pile)

Max Activating Token Index: 1005

Text #18

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.3951. Min Act: -0.0001

Data Index: 440358 (The Pile)

Max Activating Token Index: 310

Text #19

Max Range: 0.5256. Min Range: -0.5256

Max Act: 0.3805. Min Act: -0.0000

Data Index: 1316940 (The Pile)

Max Activating Token Index: 343