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

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

Neuron 1385 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6902. Min Act: -0.0001

Data Index: 823520 (C4 (Web Text))

Max Activating Token Index: 986

Text #1

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6220. Min Act: -0.0001

Data Index: 1436781 (Python Code)

Max Activating Token Index: 853

Text #2

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6525. Min Act: -0.0001

Data Index: 1577217 (Python Code)

Max Activating Token Index: 166

Text #3

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6202. Min Act: -0.0000

Data Index: 281903 (C4 (Web Text))

Max Activating Token Index: 355

Text #4

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6290. Min Act: -0.0001

Data Index: 1506817 (Python Code)

Max Activating Token Index: 917

Text #5

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6471. Min Act: -0.0001

Data Index: 1613877 (Python Code)

Max Activating Token Index: 90

Text #6

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5890. Min Act: -0.0001

Data Index: 473521 (C4 (Web Text))

Max Activating Token Index: 1017

Text #7

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6062. Min Act: -0.0001

Data Index: 429678 (C4 (Web Text))

Max Activating Token Index: 139

Text #8

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6655. Min Act: -0.0001

Data Index: 1423123 (Python Code)

Max Activating Token Index: 132

Text #9

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5656. Min Act: -0.0001

Data Index: 1416603 (Python Code)

Max Activating Token Index: 971

Text #10

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.6043. Min Act: -0.0001

Data Index: 1533631 (Python Code)

Max Activating Token Index: 324

Text #11

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5529. Min Act: -0.0001

Data Index: 1475420 (Python Code)

Max Activating Token Index: 45

Text #12

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5150. Min Act: -0.0001

Data Index: 674541 (C4 (Web Text))

Max Activating Token Index: 809

Text #13

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5480. Min Act: -0.0001

Data Index: 75827 (C4 (Web Text))

Max Activating Token Index: 1017

Text #14

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5415. Min Act: -0.0000

Data Index: 1562731 (Python Code)

Max Activating Token Index: 485

Text #15

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5331. Min Act: -0.0001

Data Index: 1519240 (Python Code)

Max Activating Token Index: 5

Text #16

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5822. Min Act: -0.0001

Data Index: 1578934 (Python Code)

Max Activating Token Index: 105

Text #17

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5122. Min Act: -0.0000

Data Index: 1592237 (Python Code)

Max Activating Token Index: 43

Text #18

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.5272. Min Act: -0.0001

Data Index: 1643141 (Python Code)

Max Activating Token Index: 886

Text #19

Max Range: 0.6902. Min Range: -0.6902

Max Act: 0.4906. Min Act: -0.0001

Data Index: 1116670 (C4 (Web Text))

Max Activating Token Index: 735