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

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

Neuron 622 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: 7.1670. Min Range: -7.1670

Max Act: 6.8419. Min Act: -0.0001

Data Index: 724703 (C4 (Web Text))

Max Activating Token Index: 943

Text #1

Max Range: 7.1670. Min Range: -7.1670

Max Act: 7.1670. Min Act: -0.0001

Data Index: 176716 (C4 (Web Text))

Max Activating Token Index: 549

Text #2

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.8741. Min Act: -0.0001

Data Index: 1028149 (C4 (Web Text))

Max Activating Token Index: 34

Text #3

Max Range: 7.1670. Min Range: -7.1670

Max Act: 7.0030. Min Act: -0.0001

Data Index: 364374 (C4 (Web Text))

Max Activating Token Index: 548

Text #4

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.4891. Min Act: -0.0001

Data Index: 299122 (C4 (Web Text))

Max Activating Token Index: 124

Text #5

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.5637. Min Act: -0.0001

Data Index: 1115899 (C4 (Web Text))

Max Activating Token Index: 1023

Text #6

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.1880. Min Act: -0.0001

Data Index: 49457 (C4 (Web Text))

Max Activating Token Index: 871

Text #7

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.6424. Min Act: -0.0001

Data Index: 35048 (C4 (Web Text))

Max Activating Token Index: 95

Text #8

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.2100. Min Act: -0.0001

Data Index: 935502 (C4 (Web Text))

Max Activating Token Index: 383

Text #9

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.8313. Min Act: -0.0001

Data Index: 1166172 (C4 (Web Text))

Max Activating Token Index: 220

Text #10

Max Range: 7.1670. Min Range: -7.1670

Max Act: 6.0789. Min Act: -0.0001

Data Index: 487788 (C4 (Web Text))

Max Activating Token Index: 519

Text #11

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.9141. Min Act: -0.0001

Data Index: 1158756 (C4 (Web Text))

Max Activating Token Index: 108

Text #12

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.8844. Min Act: -0.0001

Data Index: 490601 (C4 (Web Text))

Max Activating Token Index: 1023

Text #13

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.9871. Min Act: -0.0001

Data Index: 479344 (C4 (Web Text))

Max Activating Token Index: 259

Text #14

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.7708. Min Act: -0.0001

Data Index: 805379 (C4 (Web Text))

Max Activating Token Index: 204

Text #15

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.7639. Min Act: -0.0001

Data Index: 35631 (C4 (Web Text))

Max Activating Token Index: 336

Text #16

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.7494. Min Act: -0.0001

Data Index: 110658 (C4 (Web Text))

Max Activating Token Index: 526

Text #17

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.5919. Min Act: -0.0001

Data Index: 1260324 (C4 (Web Text))

Max Activating Token Index: 150

Text #18

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.1835. Min Act: -0.0001

Data Index: 651726 (C4 (Web Text))

Max Activating Token Index: 565

Text #19

Max Range: 7.1670. Min Range: -7.1670

Max Act: 5.6885. Min Act: -0.0001

Data Index: 838003 (C4 (Web Text))

Max Activating Token Index: 1002