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

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

Neuron 1516 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.5939. Min Act: -0.0001

Data Index: 608744 (C4 (Web Text))

Max Activating Token Index: 835

Text #1

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.6070. Min Act: -0.0001

Data Index: 218749 (C4 (Web Text))

Max Activating Token Index: 588

Text #2

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.5567. Min Act: -0.0001

Data Index: 1159847 (C4 (Web Text))

Max Activating Token Index: 988

Text #3

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.5513. Min Act: -0.0001

Data Index: 613511 (C4 (Web Text))

Max Activating Token Index: 930

Text #4

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.5097. Min Act: -0.0001

Data Index: 545141 (C4 (Web Text))

Max Activating Token Index: 461

Text #5

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.4366. Min Act: -0.0001

Data Index: 1595113 (Python Code)

Max Activating Token Index: 965

Text #6

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.4098. Min Act: -0.0001

Data Index: 1178878 (C4 (Web Text))

Max Activating Token Index: 438

Text #7

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.4165. Min Act: -0.0001

Data Index: 526487 (C4 (Web Text))

Max Activating Token Index: 29

Text #8

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.4022. Min Act: -0.0001

Data Index: 441660 (C4 (Web Text))

Max Activating Token Index: 825

Text #9

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3822. Min Act: -0.0001

Data Index: 1131126 (C4 (Web Text))

Max Activating Token Index: 312

Text #10

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3892. Min Act: -0.0001

Data Index: 90457 (C4 (Web Text))

Max Activating Token Index: 34

Text #11

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.4012. Min Act: -0.0001

Data Index: 1519663 (Python Code)

Max Activating Token Index: 612

Text #12

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3770. Min Act: -0.0001

Data Index: 1627445 (Python Code)

Max Activating Token Index: 414

Text #13

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3724. Min Act: -0.0001

Data Index: 1468044 (Python Code)

Max Activating Token Index: 633

Text #14

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3543. Min Act: -0.0001

Data Index: 1619158 (Python Code)

Max Activating Token Index: 341

Text #15

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3697. Min Act: -0.0001

Data Index: 1382834 (Python Code)

Max Activating Token Index: 734

Text #16

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3655. Min Act: -0.0001

Data Index: 1446758 (Python Code)

Max Activating Token Index: 754

Text #17

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3778. Min Act: -0.0001

Data Index: 1032965 (C4 (Web Text))

Max Activating Token Index: 533

Text #18

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3521. Min Act: -0.0001

Data Index: 548889 (C4 (Web Text))

Max Activating Token Index: 338

Text #19

Max Range: 0.6070. Min Range: -0.6070

Max Act: 0.3508. Min Act: -0.0001

Data Index: 1578862 (Python Code)

Max Activating Token Index: 980