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

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 2097 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.7617. Min Act: 0.0000

Data Index: 394618 (C4 (Web Text))

Max Activating Token Index: 880

Text #1

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.5553. Min Act: -0.0000

Data Index: 422671 (C4 (Web Text))

Max Activating Token Index: 89

Text #2

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.4868. Min Act: -0.0000

Data Index: 11855 (C4 (Web Text))

Max Activating Token Index: 289

Text #3

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.3068. Min Act: -0.0000

Data Index: 983824 (C4 (Web Text))

Max Activating Token Index: 352

Text #4

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.2810. Min Act: -0.0000

Data Index: 482802 (C4 (Web Text))

Max Activating Token Index: 780

Text #5

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.2556. Min Act: -0.0000

Data Index: 866708 (C4 (Web Text))

Max Activating Token Index: 641

Text #6

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.2417. Min Act: -0.0000

Data Index: 1140125 (C4 (Web Text))

Max Activating Token Index: 623

Text #7

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.2208. Min Act: -0.0000

Data Index: 213562 (C4 (Web Text))

Max Activating Token Index: 289

Text #8

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.1078. Min Act: 0.0000

Data Index: 963421 (C4 (Web Text))

Max Activating Token Index: 470

Text #9

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.2607. Min Act: -0.0000

Data Index: 780180 (C4 (Web Text))

Max Activating Token Index: 746

Text #10

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.1471. Min Act: -0.0000

Data Index: 1517326 (Python Code)

Max Activating Token Index: 524

Text #11

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.1448. Min Act: -0.0000

Data Index: 79221 (C4 (Web Text))

Max Activating Token Index: 1000

Text #12

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.0605. Min Act: -0.0000

Data Index: 1514521 (Python Code)

Max Activating Token Index: 107

Text #13

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.0590. Min Act: -0.0000

Data Index: 29934 (C4 (Web Text))

Max Activating Token Index: 759

Text #14

Max Range: 1.7617. Min Range: -1.7617

Max Act: 0.9458. Min Act: -0.0000

Data Index: 611270 (C4 (Web Text))

Max Activating Token Index: 345

Text #15

Max Range: 1.7617. Min Range: -1.7617

Max Act: 1.0244. Min Act: -0.0000

Data Index: 783371 (C4 (Web Text))

Max Activating Token Index: 722

Text #16

Max Range: 1.7617. Min Range: -1.7617

Max Act: 0.9719. Min Act: -0.0000

Data Index: 616619 (C4 (Web Text))

Max Activating Token Index: 783

Text #17

Max Range: 1.7617. Min Range: -1.7617

Max Act: 0.9603. Min Act: -0.0000

Data Index: 329734 (C4 (Web Text))

Max Activating Token Index: 324

Text #18

Max Range: 1.7617. Min Range: -1.7617

Max Act: 0.9369. Min Act: -0.0000

Data Index: 72834 (C4 (Web Text))

Max Activating Token Index: 645

Text #19

Max Range: 1.7617. Min Range: -1.7617

Max Act: 0.9150. Min Act: -0.0000

Data Index: 1165765 (C4 (Web Text))

Max Activating Token Index: 807