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

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

Neuron 2754 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.0787. Min Range: -1.0787

Max Act: 1.0578. Min Act: -0.0001

Data Index: 786338 (C4 (Web Text))

Max Activating Token Index: 174

Text #1

Max Range: 1.0787. Min Range: -1.0787

Max Act: 1.0056. Min Act: -0.0000

Data Index: 343524 (C4 (Web Text))

Max Activating Token Index: 490

Text #2

Max Range: 1.0787. Min Range: -1.0787

Max Act: 1.0787. Min Act: -0.0001

Data Index: 840919 (C4 (Web Text))

Max Activating Token Index: 223

Text #3

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.9976. Min Act: -0.0001

Data Index: 667545 (C4 (Web Text))

Max Activating Token Index: 98

Text #4

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8808. Min Act: -0.0001

Data Index: 251792 (C4 (Web Text))

Max Activating Token Index: 92

Text #5

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8108. Min Act: -0.0001

Data Index: 1535758 (Python Code)

Max Activating Token Index: 952

Text #6

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8515. Min Act: -0.0001

Data Index: 37048 (C4 (Web Text))

Max Activating Token Index: 658

Text #7

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.9013. Min Act: -0.0001

Data Index: 195820 (C4 (Web Text))

Max Activating Token Index: 18

Text #8

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8937. Min Act: -0.0001

Data Index: 1482496 (Python Code)

Max Activating Token Index: 671

Text #9

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8736. Min Act: -0.0001

Data Index: 725028 (C4 (Web Text))

Max Activating Token Index: 96

Text #10

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8651. Min Act: -0.0001

Data Index: 68251 (C4 (Web Text))

Max Activating Token Index: 292

Text #11

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8683. Min Act: -0.0001

Data Index: 923735 (C4 (Web Text))

Max Activating Token Index: 424

Text #12

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8473. Min Act: -0.0001

Data Index: 574970 (C4 (Web Text))

Max Activating Token Index: 540

Text #13

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.9290. Min Act: -0.0001

Data Index: 388881 (C4 (Web Text))

Max Activating Token Index: 262

Text #14

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8609. Min Act: -0.0001

Data Index: 1375490 (Python Code)

Max Activating Token Index: 936

Text #15

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8474. Min Act: -0.0001

Data Index: 978779 (C4 (Web Text))

Max Activating Token Index: 269

Text #16

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8709. Min Act: -0.0001

Data Index: 604571 (C4 (Web Text))

Max Activating Token Index: 130

Text #17

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8440. Min Act: -0.0001

Data Index: 459599 (C4 (Web Text))

Max Activating Token Index: 90

Text #18

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8123. Min Act: -0.0001

Data Index: 51501 (C4 (Web Text))

Max Activating Token Index: 851

Text #19

Max Range: 1.0787. Min Range: -1.0787

Max Act: 0.8369. Min Act: -0.0001

Data Index: 570688 (C4 (Web Text))

Max Activating Token Index: 842