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

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

Neuron 135 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.2753. Min Range: -4.2753

Max Act: 4.2753. Min Act: -0.0000

Data Index: 1050061 (C4 (Web Text))

Max Activating Token Index: 800

Text #1

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.7367. Min Act: -0.0000

Data Index: 1454145 (Python Code)

Max Activating Token Index: 452

Text #2

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.2810. Min Act: -0.0000

Data Index: 1544364 (Python Code)

Max Activating Token Index: 163

Text #3

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.1840. Min Act: -0.0000

Data Index: 1263094 (C4 (Web Text))

Max Activating Token Index: 298

Text #4

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.1851. Min Act: -0.0000

Data Index: 1423405 (Python Code)

Max Activating Token Index: 1018

Text #5

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.7226. Min Act: -0.0000

Data Index: 1351336 (C4 (Web Text))

Max Activating Token Index: 648

Text #6

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.1269. Min Act: -0.0000

Data Index: 1522223 (Python Code)

Max Activating Token Index: 403

Text #7

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.0632. Min Act: -0.0000

Data Index: 980426 (C4 (Web Text))

Max Activating Token Index: 779

Text #8

Max Range: 4.2753. Min Range: -4.2753

Max Act: 3.0367. Min Act: -0.0000

Data Index: 1089537 (C4 (Web Text))

Max Activating Token Index: 159

Text #9

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.8343. Min Act: -0.0000

Data Index: 1407702 (Python Code)

Max Activating Token Index: 320

Text #10

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.7176. Min Act: -0.0000

Data Index: 976260 (C4 (Web Text))

Max Activating Token Index: 851

Text #11

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.7171. Min Act: -0.0000

Data Index: 1397808 (Python Code)

Max Activating Token Index: 332

Text #12

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.8130. Min Act: -0.0000

Data Index: 1360981 (Python Code)

Max Activating Token Index: 324

Text #13

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.6148. Min Act: -0.0000

Data Index: 1426627 (Python Code)

Max Activating Token Index: 948

Text #14

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.7569. Min Act: -0.0000

Data Index: 1511860 (Python Code)

Max Activating Token Index: 396

Text #15

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.5243. Min Act: -0.0000

Data Index: 1480191 (Python Code)

Max Activating Token Index: 421

Text #16

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.8868. Min Act: -0.0000

Data Index: 330034 (C4 (Web Text))

Max Activating Token Index: 945

Text #17

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.6263. Min Act: -0.0000

Data Index: 719133 (C4 (Web Text))

Max Activating Token Index: 41

Text #18

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.4752. Min Act: -0.0000

Data Index: 1539990 (Python Code)

Max Activating Token Index: 421

Text #19

Max Range: 4.2753. Min Range: -4.2753

Max Act: 2.6376. Min Act: -0.0000

Data Index: 1371664 (Python Code)

Max Activating Token Index: 357