Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3224 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5729. Min Act: -0.0002

Data Index: 332323 (C4 (Web Text))

Max Activating Token Index: 863

Text #1

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5638. Min Act: -0.0001

Data Index: 678074 (C4 (Web Text))

Max Activating Token Index: 532

Text #2

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5901. Min Act: -0.0002

Data Index: 311756 (C4 (Web Text))

Max Activating Token Index: 584

Text #3

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5984. Min Act: -0.0001

Data Index: 1059563 (C4 (Web Text))

Max Activating Token Index: 1023

Text #4

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5953. Min Act: -0.0002

Data Index: 1192805 (C4 (Web Text))

Max Activating Token Index: 249

Text #5

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5371. Min Act: -0.0002

Data Index: 102777 (C4 (Web Text))

Max Activating Token Index: 851

Text #6

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5923. Min Act: -0.0001

Data Index: 578318 (C4 (Web Text))

Max Activating Token Index: 299

Text #7

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5456. Min Act: -0.0001

Data Index: 1515786 (Python Code)

Max Activating Token Index: 152

Text #8

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5376. Min Act: -0.0001

Data Index: 582013 (C4 (Web Text))

Max Activating Token Index: 27

Text #9

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5550. Min Act: -0.0002

Data Index: 1058405 (C4 (Web Text))

Max Activating Token Index: 114

Text #10

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5516. Min Act: -0.0001

Data Index: 4615 (C4 (Web Text))

Max Activating Token Index: 742

Text #11

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5534. Min Act: -0.0002

Data Index: 966658 (C4 (Web Text))

Max Activating Token Index: 4

Text #12

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5309. Min Act: -0.0002

Data Index: 406131 (C4 (Web Text))

Max Activating Token Index: 267

Text #13

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5111. Min Act: -0.0002

Data Index: 800570 (C4 (Web Text))

Max Activating Token Index: 810

Text #14

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5409. Min Act: -0.0002

Data Index: 1222281 (C4 (Web Text))

Max Activating Token Index: 292

Text #15

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5439. Min Act: -0.0001

Data Index: 1000736 (C4 (Web Text))

Max Activating Token Index: 3

Text #16

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5375. Min Act: -0.0002

Data Index: 35985 (C4 (Web Text))

Max Activating Token Index: 122

Text #17

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5356. Min Act: -0.0002

Data Index: 434453 (C4 (Web Text))

Max Activating Token Index: 649

Text #18

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5132. Min Act: -0.0001

Data Index: 740115 (C4 (Web Text))

Max Activating Token Index: 371

Text #19

Max Range: 0.5984. Min Range: -0.5984

Max Act: 0.5156. Min Act: -0.0002

Data Index: 207999 (C4 (Web Text))

Max Activating Token Index: 255