Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: The Pile

Neuron 4669 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l-pile')

Text #0

Max Range: 7.3570. Min Range: -7.3570

Max Act: 7.3570. Min Act: -0.0000

Data Index: 1998439 (The Pile)

Max Activating Token Index: 150

Text #1

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.9345. Min Act: -0.0000

Data Index: 1038424 (The Pile)

Max Activating Token Index: 392

Text #2

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.2045. Min Act: -0.0000

Data Index: 715809 (The Pile)

Max Activating Token Index: 85

Text #3

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.5968. Min Act: -0.0000

Data Index: 160148 (The Pile)

Max Activating Token Index: 536

Text #4

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.7692. Min Act: -0.0000

Data Index: 1936761 (The Pile)

Max Activating Token Index: 526

Text #5

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.7651. Min Act: -0.0000

Data Index: 1591012 (The Pile)

Max Activating Token Index: 449

Text #6

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.6127. Min Act: -0.0000

Data Index: 1776666 (The Pile)

Max Activating Token Index: 106

Text #7

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.5303. Min Act: -0.0000

Data Index: 1951640 (The Pile)

Max Activating Token Index: 610

Text #8

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.7628. Min Act: -0.0000

Data Index: 701552 (The Pile)

Max Activating Token Index: 201

Text #9

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.3451. Min Act: -0.0000

Data Index: 659690 (The Pile)

Max Activating Token Index: 438

Text #10

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.8491. Min Act: -0.0000

Data Index: 370350 (The Pile)

Max Activating Token Index: 247

Text #11

Max Range: 7.3570. Min Range: -7.3570

Max Act: 4.9125. Min Act: -0.0000

Data Index: 1192002 (The Pile)

Max Activating Token Index: 119

Text #12

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.4020. Min Act: -0.0000

Data Index: 1062275 (The Pile)

Max Activating Token Index: 366

Text #13

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.7893. Min Act: -0.0000

Data Index: 1705083 (The Pile)

Max Activating Token Index: 188

Text #14

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.1785. Min Act: -0.0000

Data Index: 1249408 (The Pile)

Max Activating Token Index: 170

Text #15

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.5342. Min Act: -0.0000

Data Index: 1227586 (The Pile)

Max Activating Token Index: 122

Text #16

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.0899. Min Act: -0.0000

Data Index: 477102 (The Pile)

Max Activating Token Index: 371

Text #17

Max Range: 7.3570. Min Range: -7.3570

Max Act: 6.5263. Min Act: -0.0000

Data Index: 922932 (The Pile)

Max Activating Token Index: 457

Text #18

Max Range: 7.3570. Min Range: -7.3570

Max Act: 5.9019. Min Act: -0.0000

Data Index: 689362 (The Pile)

Max Activating Token Index: 496

Text #19

Max Range: 7.3570. Min Range: -7.3570

Max Act: 5.2594. Min Act: -0.0000

Data Index: 89639 (The Pile)

Max Activating Token Index: 453