Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 391 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4322. Min Act: -0.0001

Data Index: 239497 (C4 (Web Text))

Max Activating Token Index: 875

Text #1

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4519. Min Act: -0.0001

Data Index: 917463 (C4 (Web Text))

Max Activating Token Index: 725

Text #2

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4792. Min Act: -0.0001

Data Index: 464137 (C4 (Web Text))

Max Activating Token Index: 46

Text #3

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3650. Min Act: -0.0001

Data Index: 954823 (C4 (Web Text))

Max Activating Token Index: 522

Text #4

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4275. Min Act: -0.0001

Data Index: 547170 (C4 (Web Text))

Max Activating Token Index: 436

Text #5

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3745. Min Act: -0.0001

Data Index: 346033 (C4 (Web Text))

Max Activating Token Index: 327

Text #6

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4187. Min Act: -0.0001

Data Index: 1017584 (C4 (Web Text))

Max Activating Token Index: 471

Text #7

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4063. Min Act: -0.0001

Data Index: 350171 (C4 (Web Text))

Max Activating Token Index: 955

Text #8

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.5064. Min Act: -0.0001

Data Index: 84308 (C4 (Web Text))

Max Activating Token Index: 291

Text #9

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4477. Min Act: -0.0001

Data Index: 604549 (C4 (Web Text))

Max Activating Token Index: 766

Text #10

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4074. Min Act: -0.0001

Data Index: 52385 (C4 (Web Text))

Max Activating Token Index: 491

Text #11

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4003. Min Act: -0.0001

Data Index: 157919 (C4 (Web Text))

Max Activating Token Index: 812

Text #12

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4199. Min Act: -0.0001

Data Index: 74782 (C4 (Web Text))

Max Activating Token Index: 910

Text #13

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4544. Min Act: -0.0001

Data Index: 916628 (C4 (Web Text))

Max Activating Token Index: 34

Text #14

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3798. Min Act: -0.0001

Data Index: 626017 (C4 (Web Text))

Max Activating Token Index: 969

Text #15

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3917. Min Act: -0.0001

Data Index: 229646 (C4 (Web Text))

Max Activating Token Index: 217

Text #16

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3793. Min Act: -0.0001

Data Index: 681635 (C4 (Web Text))

Max Activating Token Index: 264

Text #17

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3914. Min Act: -0.0001

Data Index: 149222 (C4 (Web Text))

Max Activating Token Index: 445

Text #18

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.4034. Min Act: -0.0001

Data Index: 876300 (C4 (Web Text))

Max Activating Token Index: 524

Text #19

Max Range: 0.5064. Min Range: -0.5064

Max Act: 0.3431. Min Act: -0.0001

Data Index: 1060380 (C4 (Web Text))

Max Activating Token Index: 57