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

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

Neuron 5068 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: 3.3397. Min Range: -3.3397

Max Act: 3.3013. Min Act: -0.0000

Data Index: 1234130 (C4 (Web Text))

Max Activating Token Index: 192

Text #1

Max Range: 3.3397. Min Range: -3.3397

Max Act: 3.3397. Min Act: -0.0000

Data Index: 1269966 (C4 (Web Text))

Max Activating Token Index: 389

Text #2

Max Range: 3.3397. Min Range: -3.3397

Max Act: 3.1783. Min Act: -0.0000

Data Index: 1441226 (Python Code)

Max Activating Token Index: 313

Text #3

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.9293. Min Act: -0.0000

Data Index: 1415118 (Python Code)

Max Activating Token Index: 1014

Text #4

Max Range: 3.3397. Min Range: -3.3397

Max Act: 3.0325. Min Act: -0.0000

Data Index: 623115 (C4 (Web Text))

Max Activating Token Index: 788

Text #5

Max Range: 3.3397. Min Range: -3.3397

Max Act: 3.1602. Min Act: -0.0000

Data Index: 460897 (C4 (Web Text))

Max Activating Token Index: 641

Text #6

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.7922. Min Act: -0.0000

Data Index: 667194 (C4 (Web Text))

Max Activating Token Index: 493

Text #7

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.7522. Min Act: -0.0000

Data Index: 661243 (C4 (Web Text))

Max Activating Token Index: 385

Text #8

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6953. Min Act: -0.0000

Data Index: 336398 (C4 (Web Text))

Max Activating Token Index: 817

Text #9

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6198. Min Act: -0.0000

Data Index: 442896 (C4 (Web Text))

Max Activating Token Index: 82

Text #10

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.7515. Min Act: -0.0000

Data Index: 372328 (C4 (Web Text))

Max Activating Token Index: 1019

Text #11

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6161. Min Act: -0.0000

Data Index: 1075413 (C4 (Web Text))

Max Activating Token Index: 134

Text #12

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6212. Min Act: -0.0000

Data Index: 1654180 (Python Code)

Max Activating Token Index: 1001

Text #13

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.5751. Min Act: -0.0000

Data Index: 801993 (C4 (Web Text))

Max Activating Token Index: 156

Text #14

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.5734. Min Act: -0.0000

Data Index: 80589 (C4 (Web Text))

Max Activating Token Index: 387

Text #15

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6526. Min Act: -0.0000

Data Index: 1253890 (C4 (Web Text))

Max Activating Token Index: 697

Text #16

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.5362. Min Act: -0.0000

Data Index: 1339427 (C4 (Web Text))

Max Activating Token Index: 55

Text #17

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.6507. Min Act: -0.0000

Data Index: 973974 (C4 (Web Text))

Max Activating Token Index: 945

Text #18

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.4829. Min Act: -0.0000

Data Index: 1064715 (C4 (Web Text))

Max Activating Token Index: 892

Text #19

Max Range: 3.3397. Min Range: -3.3397

Max Act: 2.4520. Min Act: -0.0000

Data Index: 163951 (C4 (Web Text))

Max Activating Token Index: 608