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

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

Neuron 3892 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0716. Min Act: -0.0000

Data Index: 825146 (C4 (Web Text))

Max Activating Token Index: 277

Text #1

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0603. Min Act: -0.0000

Data Index: 666857 (C4 (Web Text))

Max Activating Token Index: 756

Text #2

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0611. Min Act: -0.0000

Data Index: 941734 (C4 (Web Text))

Max Activating Token Index: 740

Text #3

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0554. Min Act: -0.0000

Data Index: 928461 (C4 (Web Text))

Max Activating Token Index: 1000

Text #4

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0548. Min Act: -0.0000

Data Index: 376823 (C4 (Web Text))

Max Activating Token Index: 353

Text #5

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0515. Min Act: -0.0000

Data Index: 1217556 (C4 (Web Text))

Max Activating Token Index: 921

Text #6

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0535. Min Act: -0.0000

Data Index: 1539417 (Python Code)

Max Activating Token Index: 110

Text #7

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0487. Min Act: -0.0000

Data Index: 211199 (C4 (Web Text))

Max Activating Token Index: 439

Text #8

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0494. Min Act: -0.0000

Data Index: 1464082 (Python Code)

Max Activating Token Index: 202

Text #9

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0455. Min Act: -0.0000

Data Index: 582948 (C4 (Web Text))

Max Activating Token Index: 777

Text #10

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0472. Min Act: -0.0000

Data Index: 1082422 (C4 (Web Text))

Max Activating Token Index: 197

Text #11

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0486. Min Act: -0.0000

Data Index: 1357240 (C4 (Web Text))

Max Activating Token Index: 39

Text #12

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0488. Min Act: -0.0000

Data Index: 631969 (C4 (Web Text))

Max Activating Token Index: 43

Text #13

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0476. Min Act: -0.0000

Data Index: 475259 (C4 (Web Text))

Max Activating Token Index: 249

Text #14

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0476. Min Act: -0.0000

Data Index: 1621849 (Python Code)

Max Activating Token Index: 505

Text #15

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0435. Min Act: -0.0000

Data Index: 996892 (C4 (Web Text))

Max Activating Token Index: 298

Text #16

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0492. Min Act: -0.0000

Data Index: 1528576 (Python Code)

Max Activating Token Index: 865

Text #17

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0487. Min Act: -0.0000

Data Index: 1050483 (C4 (Web Text))

Max Activating Token Index: 860

Text #18

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0488. Min Act: -0.0000

Data Index: 631921 (C4 (Web Text))

Max Activating Token Index: 284

Text #19

Max Range: 0.0716. Min Range: -0.0716

Max Act: 0.0451. Min Act: -0.0000

Data Index: 395410 (C4 (Web Text))

Max Activating Token Index: 237