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

Dataset: The Pile

Neuron 5082 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 7.3731. Min Range: -7.3731

Max Act: 7.3731. Min Act: -0.0000

Data Index: 239304 (The Pile)

Max Activating Token Index: 497

Text #1

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.5677. Min Act: -0.0000

Data Index: 1394071 (The Pile)

Max Activating Token Index: 963

Text #2

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.9769. Min Act: -0.0000

Data Index: 1375440 (The Pile)

Max Activating Token Index: 430

Text #3

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.1186. Min Act: -0.0000

Data Index: 1517789 (The Pile)

Max Activating Token Index: 817

Text #4

Max Range: 7.3731. Min Range: -7.3731

Max Act: 7.0548. Min Act: -0.0000

Data Index: 1597086 (The Pile)

Max Activating Token Index: 892

Text #5

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.8211. Min Act: -0.0000

Data Index: 172845 (The Pile)

Max Activating Token Index: 816

Text #6

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.2452. Min Act: -0.0000

Data Index: 71908 (The Pile)

Max Activating Token Index: 733

Text #7

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.2303. Min Act: -0.0000

Data Index: 1356756 (The Pile)

Max Activating Token Index: 618

Text #8

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.5854. Min Act: -0.0000

Data Index: 594981 (The Pile)

Max Activating Token Index: 115

Text #9

Max Range: 7.3731. Min Range: -7.3731

Max Act: 5.9624. Min Act: -0.0000

Data Index: 700322 (The Pile)

Max Activating Token Index: 622

Text #10

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.3277. Min Act: -0.0000

Data Index: 719091 (The Pile)

Max Activating Token Index: 563

Text #11

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.1416. Min Act: -0.0000

Data Index: 742009 (The Pile)

Max Activating Token Index: 201

Text #12

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.1043. Min Act: -0.0000

Data Index: 1278106 (The Pile)

Max Activating Token Index: 672

Text #13

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.0601. Min Act: -0.0000

Data Index: 544567 (The Pile)

Max Activating Token Index: 674

Text #14

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.2574. Min Act: -0.0000

Data Index: 750715 (The Pile)

Max Activating Token Index: 492

Text #15

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.2575. Min Act: -0.0000

Data Index: 973168 (The Pile)

Max Activating Token Index: 840

Text #16

Max Range: 7.3731. Min Range: -7.3731

Max Act: 5.9802. Min Act: 0.0000

Data Index: 1631733 (The Pile)

Max Activating Token Index: 433

Text #17

Max Range: 7.3731. Min Range: -7.3731

Max Act: 5.9762. Min Act: -0.0000

Data Index: 1244764 (The Pile)

Max Activating Token Index: 895

Text #18

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.1542. Min Act: -0.0000

Data Index: 1717043 (The Pile)

Max Activating Token Index: 325

Text #19

Max Range: 7.3731. Min Range: -7.3731

Max Act: 6.3392. Min Act: -0.0000

Data Index: 161722 (The Pile)

Max Activating Token Index: 70