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

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

Neuron 4454 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: 0.7932. Min Range: -0.7932

Max Act: 0.7827. Min Act: -0.0000

Data Index: 1074154 (C4 (Web Text))

Max Activating Token Index: 429

Text #1

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.7932. Min Act: -0.0000

Data Index: 857220 (C4 (Web Text))

Max Activating Token Index: 322

Text #2

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.7461. Min Act: -0.0000

Data Index: 245493 (C4 (Web Text))

Max Activating Token Index: 205

Text #3

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.7424. Min Act: -0.0000

Data Index: 982353 (C4 (Web Text))

Max Activating Token Index: 963

Text #4

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.6399. Min Act: -0.0000

Data Index: 1111165 (C4 (Web Text))

Max Activating Token Index: 168

Text #5

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.5579. Min Act: -0.0000

Data Index: 426601 (C4 (Web Text))

Max Activating Token Index: 732

Text #6

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.6310. Min Act: -0.0000

Data Index: 288761 (C4 (Web Text))

Max Activating Token Index: 344

Text #7

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.5013. Min Act: -0.0000

Data Index: 346313 (C4 (Web Text))

Max Activating Token Index: 601

Text #8

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.5016. Min Act: -0.0000

Data Index: 1656730 (Python Code)

Max Activating Token Index: 186

Text #9

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4844. Min Act: -0.0000

Data Index: 82067 (C4 (Web Text))

Max Activating Token Index: 997

Text #10

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4506. Min Act: -0.0000

Data Index: 880316 (C4 (Web Text))

Max Activating Token Index: 888

Text #11

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4487. Min Act: -0.0000

Data Index: 456854 (C4 (Web Text))

Max Activating Token Index: 837

Text #12

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4970. Min Act: -0.0000

Data Index: 518300 (C4 (Web Text))

Max Activating Token Index: 757

Text #13

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4195. Min Act: -0.0000

Data Index: 612927 (C4 (Web Text))

Max Activating Token Index: 839

Text #14

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4615. Min Act: -0.0000

Data Index: 445499 (C4 (Web Text))

Max Activating Token Index: 843

Text #15

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4120. Min Act: -0.0000

Data Index: 613715 (C4 (Web Text))

Max Activating Token Index: 171

Text #16

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4063. Min Act: -0.0000

Data Index: 1085518 (C4 (Web Text))

Max Activating Token Index: 246

Text #17

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.3934. Min Act: -0.0000

Data Index: 819017 (C4 (Web Text))

Max Activating Token Index: 577

Text #18

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4126. Min Act: -0.0000

Data Index: 1222049 (C4 (Web Text))

Max Activating Token Index: 838

Text #19

Max Range: 0.7932. Min Range: -0.7932

Max Act: 0.4098. Min Act: -0.0000

Data Index: 436256 (C4 (Web Text))

Max Activating Token Index: 397