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

Dataset: The Pile

Neuron 2571 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: 7.5131. Min Range: -7.5131

Max Act: 7.3422. Min Act: -0.0000

Data Index: 951330 (The Pile)

Max Activating Token Index: 675

Text #1

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.5131. Min Act: -0.0000

Data Index: 706716 (The Pile)

Max Activating Token Index: 15

Text #2

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3757. Min Act: -0.0000

Data Index: 162555 (The Pile)

Max Activating Token Index: 13

Text #3

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3703. Min Act: -0.0000

Data Index: 1505519 (The Pile)

Max Activating Token Index: 94

Text #4

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.1078. Min Act: -0.0000

Data Index: 1034203 (The Pile)

Max Activating Token Index: 108

Text #5

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3084. Min Act: -0.0000

Data Index: 62477 (The Pile)

Max Activating Token Index: 6

Text #6

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3084. Min Act: -0.0000

Data Index: 211283 (The Pile)

Max Activating Token Index: 6

Text #7

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.2351. Min Act: -0.0000

Data Index: 666951 (The Pile)

Max Activating Token Index: 179

Text #8

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3926. Min Act: -0.0000

Data Index: 369278 (The Pile)

Max Activating Token Index: 43

Text #9

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3797. Min Act: -0.0000

Data Index: 822713 (The Pile)

Max Activating Token Index: 18

Text #10

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3595. Min Act: -0.0000

Data Index: 280823 (The Pile)

Max Activating Token Index: 51

Text #11

Max Range: 7.5131. Min Range: -7.5131

Max Act: 6.8052. Min Act: -0.0000

Data Index: 453578 (The Pile)

Max Activating Token Index: 474

Text #12

Max Range: 7.5131. Min Range: -7.5131

Max Act: 6.7869. Min Act: -0.0000

Data Index: 1098418 (The Pile)

Max Activating Token Index: 121

Text #13

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.1546. Min Act: -0.0000

Data Index: 1394025 (The Pile)

Max Activating Token Index: 7

Text #14

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.1272. Min Act: -0.0000

Data Index: 935595 (The Pile)

Max Activating Token Index: 13

Text #15

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.3204. Min Act: -0.0000

Data Index: 331024 (The Pile)

Max Activating Token Index: 101

Text #16

Max Range: 7.5131. Min Range: -7.5131

Max Act: 6.5704. Min Act: -0.0000

Data Index: 429747 (The Pile)

Max Activating Token Index: 94

Text #17

Max Range: 7.5131. Min Range: -7.5131

Max Act: 6.9658. Min Act: -0.0000

Data Index: 617157 (The Pile)

Max Activating Token Index: 74

Text #18

Max Range: 7.5131. Min Range: -7.5131

Max Act: 7.1765. Min Act: -0.0000

Data Index: 1209767 (The Pile)

Max Activating Token Index: 56

Text #19

Max Range: 7.5131. Min Range: -7.5131

Max Act: 6.8794. Min Act: -0.0000

Data Index: 632695 (The Pile)

Max Activating Token Index: 71