Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 1675 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 1.0111. Min Range: -1.0111

Max Act: 1.0111. Min Act: -0.0000

Data Index: 931911 (C4 (Web Text))

Max Activating Token Index: 993

Text #1

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9595. Min Act: -0.0000

Data Index: 762253 (C4 (Web Text))

Max Activating Token Index: 312

Text #2

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9493. Min Act: -0.0000

Data Index: 1106317 (C4 (Web Text))

Max Activating Token Index: 384

Text #3

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9657. Min Act: -0.0000

Data Index: 153777 (C4 (Web Text))

Max Activating Token Index: 675

Text #4

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8898. Min Act: -0.0000

Data Index: 1001717 (C4 (Web Text))

Max Activating Token Index: 842

Text #5

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9078. Min Act: -0.0000

Data Index: 396317 (C4 (Web Text))

Max Activating Token Index: 374

Text #6

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8424. Min Act: -0.0000

Data Index: 1359973 (Python Code)

Max Activating Token Index: 637

Text #7

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.7717. Min Act: -0.0000

Data Index: 1554424 (Python Code)

Max Activating Token Index: 829

Text #8

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8955. Min Act: -0.0000

Data Index: 1301324 (C4 (Web Text))

Max Activating Token Index: 395

Text #9

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9245. Min Act: -0.0000

Data Index: 1576281 (Python Code)

Max Activating Token Index: 746

Text #10

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8728. Min Act: 0.0000

Data Index: 438079 (C4 (Web Text))

Max Activating Token Index: 446

Text #11

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8612. Min Act: -0.0000

Data Index: 1466484 (Python Code)

Max Activating Token Index: 451

Text #12

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8610. Min Act: -0.0000

Data Index: 1401162 (Python Code)

Max Activating Token Index: 665

Text #13

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.9025. Min Act: 0.0001

Data Index: 32986 (C4 (Web Text))

Max Activating Token Index: 113

Text #14

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8063. Min Act: -0.0000

Data Index: 1375318 (Python Code)

Max Activating Token Index: 549

Text #15

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8364. Min Act: -0.0000

Data Index: 540255 (C4 (Web Text))

Max Activating Token Index: 723

Text #16

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8310. Min Act: -0.0000

Data Index: 524721 (C4 (Web Text))

Max Activating Token Index: 480

Text #17

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8288. Min Act: -0.0000

Data Index: 1031075 (C4 (Web Text))

Max Activating Token Index: 591

Text #18

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8228. Min Act: -0.0000

Data Index: 516067 (C4 (Web Text))

Max Activating Token Index: 480

Text #19

Max Range: 1.0111. Min Range: -1.0111

Max Act: 0.8233. Min Act: -0.0000

Data Index: 6907 (C4 (Web Text))

Max Activating Token Index: 451