Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 1408 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0758. Min Act: -0.0002

Data Index: 482186 (C4 (Web Text))

Max Activating Token Index: 157

Text #1

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0590. Min Act: -0.0002

Data Index: 474586 (C4 (Web Text))

Max Activating Token Index: 125

Text #2

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0573. Min Act: -0.0002

Data Index: 1193143 (C4 (Web Text))

Max Activating Token Index: 599

Text #3

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0607. Min Act: -0.0002

Data Index: 22671 (C4 (Web Text))

Max Activating Token Index: 427

Text #4

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0597. Min Act: -0.0001

Data Index: 1211124 (C4 (Web Text))

Max Activating Token Index: 742

Text #5

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0574. Min Act: -0.0002

Data Index: 232372 (C4 (Web Text))

Max Activating Token Index: 55

Text #6

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0575. Min Act: -0.0002

Data Index: 1071757 (C4 (Web Text))

Max Activating Token Index: 257

Text #7

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0579. Min Act: -0.0002

Data Index: 656782 (C4 (Web Text))

Max Activating Token Index: 766

Text #8

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0573. Min Act: -0.0001

Data Index: 145335 (C4 (Web Text))

Max Activating Token Index: 215

Text #9

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0551. Min Act: -0.0002

Data Index: 528041 (C4 (Web Text))

Max Activating Token Index: 999

Text #10

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0532. Min Act: -0.0002

Data Index: 757529 (C4 (Web Text))

Max Activating Token Index: 258

Text #11

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0525. Min Act: -0.0002

Data Index: 74518 (C4 (Web Text))

Max Activating Token Index: 109

Text #12

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0512. Min Act: -0.0002

Data Index: 490604 (C4 (Web Text))

Max Activating Token Index: 865

Text #13

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0521. Min Act: -0.0002

Data Index: 587603 (C4 (Web Text))

Max Activating Token Index: 744

Text #14

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0508. Min Act: -0.0001

Data Index: 628731 (C4 (Web Text))

Max Activating Token Index: 637

Text #15

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0497. Min Act: -0.0002

Data Index: 1522758 (Python Code)

Max Activating Token Index: 44

Text #16

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0506. Min Act: -0.0002

Data Index: 1281861 (C4 (Web Text))

Max Activating Token Index: 885

Text #17

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0500. Min Act: -0.0002

Data Index: 571004 (C4 (Web Text))

Max Activating Token Index: 51

Text #18

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0503. Min Act: -0.0002

Data Index: 501149 (C4 (Web Text))

Max Activating Token Index: 809

Text #19

Max Range: 0.0758. Min Range: -0.0758

Max Act: 0.0524. Min Act: -0.0002

Data Index: 856238 (C4 (Web Text))

Max Activating Token Index: 800