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

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

Neuron 4257 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.8416. Min Range: -4.8416

Max Act: 4.8416. Min Act: -0.0000

Data Index: 307652 (C4 (Web Text))

Max Activating Token Index: 457

Text #1

Max Range: 4.8416. Min Range: -4.8416

Max Act: 4.5326. Min Act: -0.0000

Data Index: 300493 (C4 (Web Text))

Max Activating Token Index: 137

Text #2

Max Range: 4.8416. Min Range: -4.8416

Max Act: 4.1075. Min Act: -0.0000

Data Index: 406641 (C4 (Web Text))

Max Activating Token Index: 1007

Text #3

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.9016. Min Act: -0.0000

Data Index: 1625043 (Python Code)

Max Activating Token Index: 493

Text #4

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.8753. Min Act: -0.0000

Data Index: 1488737 (Python Code)

Max Activating Token Index: 439

Text #5

Max Range: 4.8416. Min Range: -4.8416

Max Act: 4.0612. Min Act: -0.0000

Data Index: 1268672 (C4 (Web Text))

Max Activating Token Index: 865

Text #6

Max Range: 4.8416. Min Range: -4.8416

Max Act: 4.1725. Min Act: -0.0000

Data Index: 1012655 (C4 (Web Text))

Max Activating Token Index: 983

Text #7

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.9107. Min Act: -0.0000

Data Index: 499374 (C4 (Web Text))

Max Activating Token Index: 19

Text #8

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.7300. Min Act: -0.0000

Data Index: 1222739 (C4 (Web Text))

Max Activating Token Index: 260

Text #9

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.8429. Min Act: -0.0000

Data Index: 81826 (C4 (Web Text))

Max Activating Token Index: 524

Text #10

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.6976. Min Act: -0.0000

Data Index: 36420 (C4 (Web Text))

Max Activating Token Index: 913

Text #11

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.9845. Min Act: -0.0000

Data Index: 416299 (C4 (Web Text))

Max Activating Token Index: 440

Text #12

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.5797. Min Act: -0.0000

Data Index: 1195476 (C4 (Web Text))

Max Activating Token Index: 701

Text #13

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.9049. Min Act: -0.0000

Data Index: 1514929 (Python Code)

Max Activating Token Index: 648

Text #14

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.5236. Min Act: -0.0000

Data Index: 982965 (C4 (Web Text))

Max Activating Token Index: 952

Text #15

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.9613. Min Act: -0.0000

Data Index: 344132 (C4 (Web Text))

Max Activating Token Index: 623

Text #16

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.4914. Min Act: -0.0000

Data Index: 1601022 (Python Code)

Max Activating Token Index: 378

Text #17

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.7448. Min Act: -0.0000

Data Index: 815836 (C4 (Web Text))

Max Activating Token Index: 1004

Text #18

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.7637. Min Act: -0.0000

Data Index: 538398 (C4 (Web Text))

Max Activating Token Index: 225

Text #19

Max Range: 4.8416. Min Range: -4.8416

Max Act: 3.5618. Min Act: -0.0000

Data Index: 300261 (C4 (Web Text))

Max Activating Token Index: 207