Model: SoLU Model: 12 Layers, 6144 Neurons per Layer

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

Neuron 5194 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Hooked Transformer Loading: HookedTransformer.from_pretrained('solu-12l')

Click the toggle to see the full text!

Text #0

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0736. Min Act: -0.0002

Data Index: 688785 (C4 (Web Text))

Max Activating Token Index: 162

Text #1

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0693. Min Act: -0.0002

Data Index: 1622597 (Python Code)

Max Activating Token Index: 110

Text #2

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0659. Min Act: -0.0002

Data Index: 1215452 (C4 (Web Text))

Max Activating Token Index: 50

Text #3

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0646. Min Act: -0.0002

Data Index: 461367 (C4 (Web Text))

Max Activating Token Index: 701

Text #4

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0636. Min Act: -0.0002

Data Index: 1649321 (Python Code)

Max Activating Token Index: 237

Text #5

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0638. Min Act: -0.0002

Data Index: 775185 (C4 (Web Text))

Max Activating Token Index: 608

Text #6

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0631. Min Act: -0.0002

Data Index: 150724 (C4 (Web Text))

Max Activating Token Index: 367

Text #7

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0611. Min Act: -0.0002

Data Index: 638844 (C4 (Web Text))

Max Activating Token Index: 298

Text #8

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0612. Min Act: -0.0002

Data Index: 430329 (C4 (Web Text))

Max Activating Token Index: 887

Text #9

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0600. Min Act: -0.0002

Data Index: 1073012 (C4 (Web Text))

Max Activating Token Index: 279

Text #10

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0615. Min Act: -0.0002

Data Index: 1464906 (Python Code)

Max Activating Token Index: 239

Text #11

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0589. Min Act: -0.0002

Data Index: 866625 (C4 (Web Text))

Max Activating Token Index: 637

Text #12

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0615. Min Act: -0.0002

Data Index: 1369179 (Python Code)

Max Activating Token Index: 77

Text #13

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0593. Min Act: -0.0002

Data Index: 1204108 (C4 (Web Text))

Max Activating Token Index: 129

Text #14

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0597. Min Act: -0.0002

Data Index: 1358930 (C4 (Web Text))

Max Activating Token Index: 17

Text #15

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0597. Min Act: -0.0002

Data Index: 333047 (C4 (Web Text))

Max Activating Token Index: 15

Text #16

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0591. Min Act: -0.0002

Data Index: 1159330 (C4 (Web Text))

Max Activating Token Index: 138

Text #17

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0594. Min Act: -0.0002

Data Index: 588481 (C4 (Web Text))

Max Activating Token Index: 239

Text #18

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0598. Min Act: -0.0002

Data Index: 1193214 (C4 (Web Text))

Max Activating Token Index: 105

Text #19

Max Range: 0.0736. Min Range: -0.0736

Max Act: 0.0607. Min Act: -0.0002

Data Index: 354786 (C4 (Web Text))

Max Activating Token Index: 107