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

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

Neuron 3327 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.7893. Min Act: -0.0000

Data Index: 1541278 (Python Code)

Max Activating Token Index: 600

Text #1

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.7208. Min Act: -0.0000

Data Index: 1462370 (Python Code)

Max Activating Token Index: 241

Text #2

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.7400. Min Act: -0.0001

Data Index: 429170 (C4 (Web Text))

Max Activating Token Index: 1013

Text #3

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.7159. Min Act: -0.0001

Data Index: 374121 (C4 (Web Text))

Max Activating Token Index: 623

Text #4

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6246. Min Act: -0.0000

Data Index: 1651265 (Python Code)

Max Activating Token Index: 662

Text #5

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6668. Min Act: -0.0001

Data Index: 429704 (C4 (Web Text))

Max Activating Token Index: 131

Text #6

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6064. Min Act: -0.0001

Data Index: 1204881 (C4 (Web Text))

Max Activating Token Index: 848

Text #7

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6278. Min Act: -0.0001

Data Index: 1231577 (C4 (Web Text))

Max Activating Token Index: 696

Text #8

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5988. Min Act: -0.0000

Data Index: 936527 (C4 (Web Text))

Max Activating Token Index: 511

Text #9

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6373. Min Act: -0.0001

Data Index: 134870 (C4 (Web Text))

Max Activating Token Index: 679

Text #10

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6295. Min Act: -0.0001

Data Index: 1495863 (Python Code)

Max Activating Token Index: 749

Text #11

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.6069. Min Act: -0.0001

Data Index: 195405 (C4 (Web Text))

Max Activating Token Index: 544

Text #12

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5755. Min Act: -0.0001

Data Index: 747365 (C4 (Web Text))

Max Activating Token Index: 504

Text #13

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5774. Min Act: -0.0000

Data Index: 653536 (C4 (Web Text))

Max Activating Token Index: 62

Text #14

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5636. Min Act: -0.0001

Data Index: 1487542 (Python Code)

Max Activating Token Index: 65

Text #15

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5828. Min Act: -0.0000

Data Index: 1482223 (Python Code)

Max Activating Token Index: 295

Text #16

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5560. Min Act: -0.0000

Data Index: 1610357 (Python Code)

Max Activating Token Index: 508

Text #17

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5773. Min Act: -0.0001

Data Index: 335620 (C4 (Web Text))

Max Activating Token Index: 517

Text #18

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5666. Min Act: -0.0000

Data Index: 1395655 (Python Code)

Max Activating Token Index: 227

Text #19

Max Range: 0.7893. Min Range: -0.7893

Max Act: 0.5740. Min Act: -0.0001

Data Index: 624866 (C4 (Web Text))

Max Activating Token Index: 920