Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 3931 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 5.7757. Min Range: -5.7757

Max Act: 5.7757. Min Act: -0.0001

Data Index: 1023705 (C4 (Web Text))

Max Activating Token Index: 423

Text #1

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.9007. Min Act: -0.0001

Data Index: 920566 (C4 (Web Text))

Max Activating Token Index: 699

Text #2

Max Range: 5.7757. Min Range: -5.7757

Max Act: 5.1493. Min Act: -0.0001

Data Index: 356737 (C4 (Web Text))

Max Activating Token Index: 114

Text #3

Max Range: 5.7757. Min Range: -5.7757

Max Act: 5.1023. Min Act: -0.0001

Data Index: 257300 (C4 (Web Text))

Max Activating Token Index: 986

Text #4

Max Range: 5.7757. Min Range: -5.7757

Max Act: 5.1246. Min Act: -0.0001

Data Index: 693121 (C4 (Web Text))

Max Activating Token Index: 331

Text #5

Max Range: 5.7757. Min Range: -5.7757

Max Act: 5.1409. Min Act: -0.0001

Data Index: 1249462 (C4 (Web Text))

Max Activating Token Index: 660

Text #6

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.7262. Min Act: -0.0001

Data Index: 816946 (C4 (Web Text))

Max Activating Token Index: 1016

Text #7

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.8031. Min Act: -0.0001

Data Index: 124306 (C4 (Web Text))

Max Activating Token Index: 319

Text #8

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.6507. Min Act: -0.0001

Data Index: 210196 (C4 (Web Text))

Max Activating Token Index: 86

Text #9

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.5759. Min Act: -0.0001

Data Index: 332324 (C4 (Web Text))

Max Activating Token Index: 369

Text #10

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.5264. Min Act: -0.0001

Data Index: 453296 (C4 (Web Text))

Max Activating Token Index: 7

Text #11

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.5232. Min Act: -0.0001

Data Index: 626625 (C4 (Web Text))

Max Activating Token Index: 930

Text #12

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.3734. Min Act: -0.0001

Data Index: 383733 (C4 (Web Text))

Max Activating Token Index: 853

Text #13

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.6751. Min Act: -0.0001

Data Index: 1202000 (C4 (Web Text))

Max Activating Token Index: 925

Text #14

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.4754. Min Act: -0.0001

Data Index: 791163 (C4 (Web Text))

Max Activating Token Index: 22

Text #15

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.5873. Min Act: -0.0001

Data Index: 747772 (C4 (Web Text))

Max Activating Token Index: 117

Text #16

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.5650. Min Act: -0.0001

Data Index: 1216506 (C4 (Web Text))

Max Activating Token Index: 319

Text #17

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.3679. Min Act: -0.0001

Data Index: 371377 (C4 (Web Text))

Max Activating Token Index: 804

Text #18

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.3920. Min Act: -0.0001

Data Index: 1352323 (C4 (Web Text))

Max Activating Token Index: 188

Text #19

Max Range: 5.7757. Min Range: -5.7757

Max Act: 4.4547. Min Act: -0.0001

Data Index: 280515 (C4 (Web Text))

Max Activating Token Index: 291