Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

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

Neuron 1463 in Layer 2

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0535. Min Act: -0.0002

Data Index: 1321522 (C4 (Web Text))

Max Activating Token Index: 575

Text #1

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0510. Min Act: -0.0002

Data Index: 1636120 (Python Code)

Max Activating Token Index: 129

Text #2

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0485. Min Act: -0.0002

Data Index: 954798 (C4 (Web Text))

Max Activating Token Index: 97

Text #3

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0454. Min Act: -0.0002

Data Index: 206103 (C4 (Web Text))

Max Activating Token Index: 28

Text #4

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0444. Min Act: -0.0002

Data Index: 1243051 (C4 (Web Text))

Max Activating Token Index: 930

Text #5

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0444. Min Act: -0.0002

Data Index: 1536483 (Python Code)

Max Activating Token Index: 456

Text #6

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0451. Min Act: -0.0002

Data Index: 1209744 (C4 (Web Text))

Max Activating Token Index: 904

Text #7

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0448. Min Act: -0.0002

Data Index: 317934 (C4 (Web Text))

Max Activating Token Index: 928

Text #8

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0490. Min Act: -0.0002

Data Index: 1597429 (Python Code)

Max Activating Token Index: 150

Text #9

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0449. Min Act: -0.0002

Data Index: 119441 (C4 (Web Text))

Max Activating Token Index: 490

Text #10

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0438. Min Act: -0.0002

Data Index: 60448 (C4 (Web Text))

Max Activating Token Index: 763

Text #11

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0426. Min Act: -0.0002

Data Index: 1533752 (Python Code)

Max Activating Token Index: 987

Text #12

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0429. Min Act: -0.0002

Data Index: 1216307 (C4 (Web Text))

Max Activating Token Index: 914

Text #13

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0437. Min Act: -0.0002

Data Index: 1158243 (C4 (Web Text))

Max Activating Token Index: 675

Text #14

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0409. Min Act: -0.0002

Data Index: 1377893 (Python Code)

Max Activating Token Index: 67

Text #15

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0441. Min Act: -0.0002

Data Index: 157743 (C4 (Web Text))

Max Activating Token Index: 229

Text #16

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0412. Min Act: -0.0002

Data Index: 1036284 (C4 (Web Text))

Max Activating Token Index: 710

Text #17

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0421. Min Act: -0.0002

Data Index: 1504071 (Python Code)

Max Activating Token Index: 373

Text #18

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0439. Min Act: -0.0002

Data Index: 1085799 (C4 (Web Text))

Max Activating Token Index: 886

Text #19

Max Range: 0.0535. Min Range: -0.0535

Max Act: 0.0439. Min Act: -0.0002

Data Index: 1642184 (Python Code)

Max Activating Token Index: 615