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

Dataset: The Pile

Neuron 3494 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1164. Min Range: -1.1164

Max Act: 1.1164. Min Act: -0.0001

Data Index: 894803 (The Pile)

Max Activating Token Index: 505

Text #1

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.3535. Min Act: -0.0001

Data Index: 1632411 (The Pile)

Max Activating Token Index: 494

Text #2

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0063. Min Act: -0.0001

Data Index: 43463 (The Pile)

Max Activating Token Index: 796

Text #3

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.1386. Min Act: -0.0000

Data Index: 1624447 (The Pile)

Max Activating Token Index: 35

Text #4

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.1229. Min Act: -0.0000

Data Index: 1054547 (The Pile)

Max Activating Token Index: 276

Text #5

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.1128. Min Act: -0.0001

Data Index: 134662 (The Pile)

Max Activating Token Index: 14

Text #6

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0327. Min Act: -0.0000

Data Index: 161168 (The Pile)

Max Activating Token Index: 128

Text #7

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.1209. Min Act: -0.0000

Data Index: 1273698 (The Pile)

Max Activating Token Index: 23

Text #8

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0933. Min Act: -0.0000

Data Index: 1203927 (The Pile)

Max Activating Token Index: 369

Text #9

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0980. Min Act: -0.0000

Data Index: 596853 (The Pile)

Max Activating Token Index: 290

Text #10

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0901. Min Act: -0.0001

Data Index: 1995605 (The Pile)

Max Activating Token Index: 292

Text #11

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0800. Min Act: -0.0000

Data Index: 318669 (The Pile)

Max Activating Token Index: 465

Text #12

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0750. Min Act: -0.0001

Data Index: 611658 (The Pile)

Max Activating Token Index: 330

Text #13

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0893. Min Act: -0.0000

Data Index: 1564713 (The Pile)

Max Activating Token Index: 20

Text #14

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0802. Min Act: -0.0000

Data Index: 1617503 (The Pile)

Max Activating Token Index: 25

Text #15

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0728. Min Act: -0.0001

Data Index: 1960479 (The Pile)

Max Activating Token Index: 94

Text #16

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0701. Min Act: -0.0000

Data Index: 1416475 (The Pile)

Max Activating Token Index: 966

Text #17

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0648. Min Act: -0.0000

Data Index: 824473 (The Pile)

Max Activating Token Index: 562

Text #18

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0713. Min Act: -0.0000

Data Index: 902326 (The Pile)

Max Activating Token Index: 425

Text #19

Max Range: 1.1164. Min Range: -1.1164

Max Act: 0.0745. Min Act: -0.0000

Data Index: 529567 (The Pile)

Max Activating Token Index: 144