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

Dataset: The Pile

Neuron 603 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.5790. Min Range: -5.5790

Max Act: 5.5347. Min Act: -0.0000

Data Index: 45023 (The Pile)

Max Activating Token Index: 169

Text #1

Max Range: 5.5790. Min Range: -5.5790

Max Act: 5.5790. Min Act: -0.0000

Data Index: 1599706 (The Pile)

Max Activating Token Index: 24

Text #2

Max Range: 5.5790. Min Range: -5.5790

Max Act: 5.5554. Min Act: -0.0000

Data Index: 1746041 (The Pile)

Max Activating Token Index: 22

Text #3

Max Range: 5.5790. Min Range: -5.5790

Max Act: 5.5257. Min Act: -0.0000

Data Index: 552883 (The Pile)

Max Activating Token Index: 4

Text #4

Max Range: 5.5790. Min Range: -5.5790

Max Act: 5.2966. Min Act: -0.0000

Data Index: 1123009 (The Pile)

Max Activating Token Index: 114

Text #5

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.6899. Min Act: -0.0000

Data Index: 1391294 (The Pile)

Max Activating Token Index: 5

Text #6

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.4535. Min Act: -0.0000

Data Index: 841202 (The Pile)

Max Activating Token Index: 197

Text #7

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.2509. Min Act: -0.0000

Data Index: 1095455 (The Pile)

Max Activating Token Index: 94

Text #8

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.4310. Min Act: -0.0000

Data Index: 1894723 (The Pile)

Max Activating Token Index: 41

Text #9

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.5847. Min Act: -0.0000

Data Index: 1269573 (The Pile)

Max Activating Token Index: 39

Text #10

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.2465. Min Act: -0.0000

Data Index: 530030 (The Pile)

Max Activating Token Index: 157

Text #11

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.3711. Min Act: -0.0000

Data Index: 601494 (The Pile)

Max Activating Token Index: 175

Text #12

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.3348. Min Act: -0.0000

Data Index: 1083180 (The Pile)

Max Activating Token Index: 54

Text #13

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.1534. Min Act: -0.0000

Data Index: 689139 (The Pile)

Max Activating Token Index: 23

Text #14

Max Range: 5.5790. Min Range: -5.5790

Max Act: 3.4605. Min Act: -0.0000

Data Index: 857188 (The Pile)

Max Activating Token Index: 403

Text #15

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.0602. Min Act: -0.0000

Data Index: 847680 (The Pile)

Max Activating Token Index: 432

Text #16

Max Range: 5.5790. Min Range: -5.5790

Max Act: 3.8608. Min Act: -0.0000

Data Index: 545261 (The Pile)

Max Activating Token Index: 6

Text #17

Max Range: 5.5790. Min Range: -5.5790

Max Act: 3.4625. Min Act: -0.0000

Data Index: 1495722 (The Pile)

Max Activating Token Index: 772

Text #18

Max Range: 5.5790. Min Range: -5.5790

Max Act: 4.0711. Min Act: -0.0000

Data Index: 771887 (The Pile)

Max Activating Token Index: 389

Text #19

Max Range: 5.5790. Min Range: -5.5790

Max Act: 3.8120. Min Act: -0.0000

Data Index: 1144435 (The Pile)

Max Activating Token Index: 119