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

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

Neuron 4691 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5766. Min Act: -0.0001

Data Index: 33062 (C4 (Web Text))

Max Activating Token Index: 838

Text #1

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5919. Min Act: -0.0001

Data Index: 914282 (C4 (Web Text))

Max Activating Token Index: 740

Text #2

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5085. Min Act: -0.0001

Data Index: 1056110 (C4 (Web Text))

Max Activating Token Index: 581

Text #3

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5049. Min Act: -0.0001

Data Index: 927138 (C4 (Web Text))

Max Activating Token Index: 168

Text #4

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5044. Min Act: -0.0001

Data Index: 1031683 (C4 (Web Text))

Max Activating Token Index: 828

Text #5

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5051. Min Act: -0.0001

Data Index: 228688 (C4 (Web Text))

Max Activating Token Index: 678

Text #6

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4929. Min Act: -0.0001

Data Index: 81210 (C4 (Web Text))

Max Activating Token Index: 1000

Text #7

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4831. Min Act: -0.0001

Data Index: 1251272 (C4 (Web Text))

Max Activating Token Index: 611

Text #8

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5148. Min Act: -0.0001

Data Index: 610450 (C4 (Web Text))

Max Activating Token Index: 1017

Text #9

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4907. Min Act: -0.0001

Data Index: 852750 (C4 (Web Text))

Max Activating Token Index: 648

Text #10

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4811. Min Act: -0.0001

Data Index: 816488 (C4 (Web Text))

Max Activating Token Index: 375

Text #11

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4587. Min Act: -0.0001

Data Index: 129798 (C4 (Web Text))

Max Activating Token Index: 437

Text #12

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4684. Min Act: -0.0001

Data Index: 58262 (C4 (Web Text))

Max Activating Token Index: 984

Text #13

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4595. Min Act: -0.0001

Data Index: 723405 (C4 (Web Text))

Max Activating Token Index: 148

Text #14

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.5304. Min Act: -0.0001

Data Index: 64263 (C4 (Web Text))

Max Activating Token Index: 818

Text #15

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4734. Min Act: -0.0001

Data Index: 283991 (C4 (Web Text))

Max Activating Token Index: 166

Text #16

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4972. Min Act: -0.0001

Data Index: 1025173 (C4 (Web Text))

Max Activating Token Index: 145

Text #17

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4619. Min Act: -0.0001

Data Index: 1005485 (C4 (Web Text))

Max Activating Token Index: 869

Text #18

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4421. Min Act: -0.0001

Data Index: 611379 (C4 (Web Text))

Max Activating Token Index: 569

Text #19

Max Range: 0.5919. Min Range: -0.5919

Max Act: 0.4355. Min Act: -0.0001

Data Index: 926785 (C4 (Web Text))

Max Activating Token Index: 830