Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 191 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1741. Min Act: -0.0001

Data Index: 527780 (C4 (Web Text))

Max Activating Token Index: 993

Text #1

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1768. Min Act: -0.0001

Data Index: 212277 (C4 (Web Text))

Max Activating Token Index: 397

Text #2

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1762. Min Act: -0.0001

Data Index: 132902 (C4 (Web Text))

Max Activating Token Index: 965

Text #3

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1681. Min Act: -0.0001

Data Index: 1266869 (C4 (Web Text))

Max Activating Token Index: 197

Text #4

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1700. Min Act: -0.0001

Data Index: 243734 (C4 (Web Text))

Max Activating Token Index: 50

Text #5

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1516. Min Act: -0.0001

Data Index: 234904 (C4 (Web Text))

Max Activating Token Index: 184

Text #6

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1526. Min Act: -0.0001

Data Index: 414855 (C4 (Web Text))

Max Activating Token Index: 1001

Text #7

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1521. Min Act: -0.0001

Data Index: 72649 (C4 (Web Text))

Max Activating Token Index: 311

Text #8

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1551. Min Act: -0.0001

Data Index: 388103 (C4 (Web Text))

Max Activating Token Index: 92

Text #9

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1645. Min Act: -0.0001

Data Index: 555159 (C4 (Web Text))

Max Activating Token Index: 819

Text #10

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1538. Min Act: -0.0001

Data Index: 998050 (C4 (Web Text))

Max Activating Token Index: 685

Text #11

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1590. Min Act: -0.0001

Data Index: 466062 (C4 (Web Text))

Max Activating Token Index: 32

Text #12

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1666. Min Act: -0.0001

Data Index: 805056 (C4 (Web Text))

Max Activating Token Index: 519

Text #13

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1418. Min Act: -0.0001

Data Index: 1109267 (C4 (Web Text))

Max Activating Token Index: 127

Text #14

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1579. Min Act: -0.0001

Data Index: 761232 (C4 (Web Text))

Max Activating Token Index: 24

Text #15

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1470. Min Act: -0.0001

Data Index: 419686 (C4 (Web Text))

Max Activating Token Index: 183

Text #16

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1420. Min Act: -0.0001

Data Index: 925557 (C4 (Web Text))

Max Activating Token Index: 65

Text #17

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1567. Min Act: -0.0001

Data Index: 423188 (C4 (Web Text))

Max Activating Token Index: 55

Text #18

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1433. Min Act: -0.0001

Data Index: 1154643 (C4 (Web Text))

Max Activating Token Index: 828

Text #19

Max Range: 0.1768. Min Range: -0.1768

Max Act: 0.1602. Min Act: -0.0001

Data Index: 1244014 (C4 (Web Text))

Max Activating Token Index: 268