Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3688 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1866. Min Act: -0.0001

Data Index: 746172 (C4 (Web Text))

Max Activating Token Index: 23

Text #1

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1691. Min Act: -0.0001

Data Index: 1156676 (C4 (Web Text))

Max Activating Token Index: 764

Text #2

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1727. Min Act: -0.0001

Data Index: 1288140 (C4 (Web Text))

Max Activating Token Index: 920

Text #3

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1668. Min Act: -0.0001

Data Index: 512636 (C4 (Web Text))

Max Activating Token Index: 935

Text #4

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1774. Min Act: -0.0001

Data Index: 1013819 (C4 (Web Text))

Max Activating Token Index: 812

Text #5

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1817. Min Act: -0.0001

Data Index: 719654 (C4 (Web Text))

Max Activating Token Index: 796

Text #6

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1759. Min Act: -0.0001

Data Index: 4801 (C4 (Web Text))

Max Activating Token Index: 828

Text #7

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1571. Min Act: -0.0001

Data Index: 1041713 (C4 (Web Text))

Max Activating Token Index: 333

Text #8

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1622. Min Act: -0.0001

Data Index: 729709 (C4 (Web Text))

Max Activating Token Index: 886

Text #9

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1618. Min Act: -0.0001

Data Index: 132128 (C4 (Web Text))

Max Activating Token Index: 524

Text #10

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1543. Min Act: -0.0001

Data Index: 423501 (C4 (Web Text))

Max Activating Token Index: 150

Text #11

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1541. Min Act: -0.0001

Data Index: 1625817 (Python Code)

Max Activating Token Index: 311

Text #12

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1474. Min Act: -0.0001

Data Index: 837588 (C4 (Web Text))

Max Activating Token Index: 911

Text #13

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1477. Min Act: -0.0001

Data Index: 961109 (C4 (Web Text))

Max Activating Token Index: 380

Text #14

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1531. Min Act: -0.0001

Data Index: 577521 (C4 (Web Text))

Max Activating Token Index: 773

Text #15

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1567. Min Act: -0.0001

Data Index: 476529 (C4 (Web Text))

Max Activating Token Index: 563

Text #16

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1441. Min Act: -0.0001

Data Index: 1353483 (C4 (Web Text))

Max Activating Token Index: 744

Text #17

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1491. Min Act: -0.0001

Data Index: 365776 (C4 (Web Text))

Max Activating Token Index: 913

Text #18

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1457. Min Act: -0.0001

Data Index: 1319582 (C4 (Web Text))

Max Activating Token Index: 796

Text #19

Max Range: 0.1866. Min Range: -0.1866

Max Act: 0.1528. Min Act: -0.0001

Data Index: 198274 (C4 (Web Text))

Max Activating Token Index: 711