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

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

Neuron 4450 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 5.8508. Min Range: -5.8508

Max Act: 5.8508. Min Act: -0.0000

Data Index: 271732 (C4 (Web Text))

Max Activating Token Index: 446

Text #1

Max Range: 5.8508. Min Range: -5.8508

Max Act: 4.7262. Min Act: -0.0000

Data Index: 594197 (C4 (Web Text))

Max Activating Token Index: 307

Text #2

Max Range: 5.8508. Min Range: -5.8508

Max Act: 4.2702. Min Act: -0.0000

Data Index: 1257631 (C4 (Web Text))

Max Activating Token Index: 586

Text #3

Max Range: 5.8508. Min Range: -5.8508

Max Act: 4.1837. Min Act: -0.0000

Data Index: 238093 (C4 (Web Text))

Max Activating Token Index: 730

Text #4

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.5852. Min Act: -0.0000

Data Index: 158772 (C4 (Web Text))

Max Activating Token Index: 220

Text #5

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.5341. Min Act: -0.0000

Data Index: 930728 (C4 (Web Text))

Max Activating Token Index: 952

Text #6

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.3731. Min Act: -0.0000

Data Index: 136827 (C4 (Web Text))

Max Activating Token Index: 497

Text #7

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.6645. Min Act: -0.0000

Data Index: 981596 (C4 (Web Text))

Max Activating Token Index: 252

Text #8

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.2694. Min Act: -0.0000

Data Index: 448174 (C4 (Web Text))

Max Activating Token Index: 688

Text #9

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.2527. Min Act: -0.0000

Data Index: 1237398 (C4 (Web Text))

Max Activating Token Index: 271

Text #10

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.2286. Min Act: -0.0000

Data Index: 465810 (C4 (Web Text))

Max Activating Token Index: 862

Text #11

Max Range: 5.8508. Min Range: -5.8508

Max Act: 3.1793. Min Act: -0.0000

Data Index: 659320 (C4 (Web Text))

Max Activating Token Index: 47

Text #12

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.8798. Min Act: -0.0000

Data Index: 1067870 (C4 (Web Text))

Max Activating Token Index: 950

Text #13

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.7378. Min Act: -0.0000

Data Index: 950278 (C4 (Web Text))

Max Activating Token Index: 860

Text #14

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.7207. Min Act: -0.0000

Data Index: 776243 (C4 (Web Text))

Max Activating Token Index: 725

Text #15

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.9682. Min Act: -0.0000

Data Index: 913034 (C4 (Web Text))

Max Activating Token Index: 1018

Text #16

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.6073. Min Act: -0.0000

Data Index: 425739 (C4 (Web Text))

Max Activating Token Index: 1018

Text #17

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.8366. Min Act: -0.0000

Data Index: 1128718 (C4 (Web Text))

Max Activating Token Index: 788

Text #18

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.9181. Min Act: -0.0000

Data Index: 1122381 (C4 (Web Text))

Max Activating Token Index: 601

Text #19

Max Range: 5.8508. Min Range: -5.8508

Max Act: 2.6556. Min Act: -0.0000

Data Index: 909530 (C4 (Web Text))

Max Activating Token Index: 525