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

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

Neuron 20 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 1.1523. Min Range: -1.1523

Max Act: 1.1523. Min Act: -0.0002

Data Index: 349466 (C4 (Web Text))

Max Activating Token Index: 637

Text #1

Max Range: 1.1523. Min Range: -1.1523

Max Act: 1.0097. Min Act: -0.0001

Data Index: 1045544 (C4 (Web Text))

Max Activating Token Index: 529

Text #2

Max Range: 1.1523. Min Range: -1.1523

Max Act: 1.0062. Min Act: -0.0002

Data Index: 1307713 (C4 (Web Text))

Max Activating Token Index: 662

Text #3

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9422. Min Act: -0.0001

Data Index: 565957 (C4 (Web Text))

Max Activating Token Index: 512

Text #4

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9500. Min Act: -0.0001

Data Index: 716950 (C4 (Web Text))

Max Activating Token Index: 470

Text #5

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9661. Min Act: -0.0002

Data Index: 470269 (C4 (Web Text))

Max Activating Token Index: 463

Text #6

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9083. Min Act: -0.0001

Data Index: 733352 (C4 (Web Text))

Max Activating Token Index: 224

Text #7

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9600. Min Act: -0.0001

Data Index: 640765 (C4 (Web Text))

Max Activating Token Index: 891

Text #8

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9292. Min Act: -0.0002

Data Index: 75309 (C4 (Web Text))

Max Activating Token Index: 879

Text #9

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.7812. Min Act: -0.0002

Data Index: 262023 (C4 (Web Text))

Max Activating Token Index: 220

Text #10

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8452. Min Act: -0.0001

Data Index: 698560 (C4 (Web Text))

Max Activating Token Index: 731

Text #11

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8990. Min Act: -0.0002

Data Index: 587431 (C4 (Web Text))

Max Activating Token Index: 500

Text #12

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9643. Min Act: -0.0001

Data Index: 109354 (C4 (Web Text))

Max Activating Token Index: 643

Text #13

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8869. Min Act: -0.0001

Data Index: 7798 (C4 (Web Text))

Max Activating Token Index: 520

Text #14

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.9056. Min Act: -0.0001

Data Index: 921998 (C4 (Web Text))

Max Activating Token Index: 537

Text #15

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8276. Min Act: -0.0001

Data Index: 530507 (C4 (Web Text))

Max Activating Token Index: 658

Text #16

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8303. Min Act: -0.0001

Data Index: 260060 (C4 (Web Text))

Max Activating Token Index: 604

Text #17

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.8840. Min Act: -0.0001

Data Index: 1053406 (C4 (Web Text))

Max Activating Token Index: 359

Text #18

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.7016. Min Act: -0.0002

Data Index: 85693 (C4 (Web Text))

Max Activating Token Index: 741

Text #19

Max Range: 1.1523. Min Range: -1.1523

Max Act: 0.7756. Min Act: -0.0002

Data Index: 522296 (C4 (Web Text))

Max Activating Token Index: 293