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

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

Neuron 3927 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: 1.1823. Min Range: -1.1823

Max Act: 1.1621. Min Act: -0.0001

Data Index: 1223114 (C4 (Web Text))

Max Activating Token Index: 4

Text #1

Max Range: 1.1823. Min Range: -1.1823

Max Act: 1.1823. Min Act: -0.0001

Data Index: 1102333 (C4 (Web Text))

Max Activating Token Index: 20

Text #2

Max Range: 1.1823. Min Range: -1.1823

Max Act: 1.0571. Min Act: -0.0001

Data Index: 1042233 (C4 (Web Text))

Max Activating Token Index: 84

Text #3

Max Range: 1.1823. Min Range: -1.1823

Max Act: 1.0013. Min Act: -0.0001

Data Index: 381074 (C4 (Web Text))

Max Activating Token Index: 588

Text #4

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9313. Min Act: -0.0001

Data Index: 969542 (C4 (Web Text))

Max Activating Token Index: 5

Text #5

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8944. Min Act: -0.0001

Data Index: 1235314 (C4 (Web Text))

Max Activating Token Index: 731

Text #6

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9260. Min Act: -0.0001

Data Index: 340785 (C4 (Web Text))

Max Activating Token Index: 4

Text #7

Max Range: 1.1823. Min Range: -1.1823

Max Act: 1.0095. Min Act: -0.0001

Data Index: 731548 (C4 (Web Text))

Max Activating Token Index: 834

Text #8

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9762. Min Act: -0.0001

Data Index: 347534 (C4 (Web Text))

Max Activating Token Index: 2

Text #9

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9762. Min Act: -0.0001

Data Index: 163395 (C4 (Web Text))

Max Activating Token Index: 2

Text #10

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9366. Min Act: -0.0001

Data Index: 174768 (C4 (Web Text))

Max Activating Token Index: 6

Text #11

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.9079. Min Act: -0.0001

Data Index: 705316 (C4 (Web Text))

Max Activating Token Index: 4

Text #12

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8508. Min Act: -0.0001

Data Index: 400008 (C4 (Web Text))

Max Activating Token Index: 882

Text #13

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8675. Min Act: -0.0001

Data Index: 619898 (C4 (Web Text))

Max Activating Token Index: 2

Text #14

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8675. Min Act: -0.0001

Data Index: 1150022 (C4 (Web Text))

Max Activating Token Index: 2

Text #15

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8681. Min Act: -0.0001

Data Index: 724375 (C4 (Web Text))

Max Activating Token Index: 5

Text #16

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8583. Min Act: -0.0001

Data Index: 755983 (C4 (Web Text))

Max Activating Token Index: 665

Text #17

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8021. Min Act: -0.0001

Data Index: 397720 (C4 (Web Text))

Max Activating Token Index: 383

Text #18

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.8633. Min Act: -0.0001

Data Index: 851656 (C4 (Web Text))

Max Activating Token Index: 900

Text #19

Max Range: 1.1823. Min Range: -1.1823

Max Act: 0.7857. Min Act: -0.0001

Data Index: 158100 (C4 (Web Text))

Max Activating Token Index: 633