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

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

Neuron 4218 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: 0.8424. Min Range: -0.8424

Max Act: 0.8092. Min Act: -0.0000

Data Index: 968807 (C4 (Web Text))

Max Activating Token Index: 851

Text #1

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.8249. Min Act: -0.0000

Data Index: 255539 (C4 (Web Text))

Max Activating Token Index: 538

Text #2

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.7786. Min Act: -0.0000

Data Index: 829392 (C4 (Web Text))

Max Activating Token Index: 909

Text #3

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.8424. Min Act: -0.0000

Data Index: 68937 (C4 (Web Text))

Max Activating Token Index: 122

Text #4

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.7393. Min Act: -0.0000

Data Index: 38270 (C4 (Web Text))

Max Activating Token Index: 485

Text #5

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.7825. Min Act: -0.0000

Data Index: 768321 (C4 (Web Text))

Max Activating Token Index: 365

Text #6

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.6669. Min Act: -0.0000

Data Index: 1134943 (C4 (Web Text))

Max Activating Token Index: 837

Text #7

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.6821. Min Act: -0.0000

Data Index: 888120 (C4 (Web Text))

Max Activating Token Index: 436

Text #8

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.6479. Min Act: -0.0000

Data Index: 310149 (C4 (Web Text))

Max Activating Token Index: 489

Text #9

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5893. Min Act: -0.0000

Data Index: 3686 (C4 (Web Text))

Max Activating Token Index: 868

Text #10

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5475. Min Act: -0.0000

Data Index: 1615925 (Python Code)

Max Activating Token Index: 244

Text #11

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.6378. Min Act: -0.0000

Data Index: 613504 (C4 (Web Text))

Max Activating Token Index: 934

Text #12

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5333. Min Act: -0.0000

Data Index: 989098 (C4 (Web Text))

Max Activating Token Index: 321

Text #13

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5418. Min Act: -0.0000

Data Index: 955642 (C4 (Web Text))

Max Activating Token Index: 527

Text #14

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.4932. Min Act: -0.0000

Data Index: 1046508 (C4 (Web Text))

Max Activating Token Index: 971

Text #15

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.4989. Min Act: -0.0000

Data Index: 340712 (C4 (Web Text))

Max Activating Token Index: 450

Text #16

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5147. Min Act: -0.0000

Data Index: 302986 (C4 (Web Text))

Max Activating Token Index: 907

Text #17

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.5148. Min Act: -0.0000

Data Index: 381810 (C4 (Web Text))

Max Activating Token Index: 310

Text #18

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.4892. Min Act: -0.0000

Data Index: 1476829 (Python Code)

Max Activating Token Index: 913

Text #19

Max Range: 0.8424. Min Range: -0.8424

Max Act: 0.4765. Min Act: -0.0000

Data Index: 967222 (C4 (Web Text))

Max Activating Token Index: 497