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

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

Neuron 2015 in Layer 5

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.5648. Min Act: -0.0001

Data Index: 1504946 (Python Code)

Max Activating Token Index: 933

Text #1

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3996. Min Act: -0.0001

Data Index: 1516106 (Python Code)

Max Activating Token Index: 52

Text #2

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3985. Min Act: -0.0001

Data Index: 910545 (C4 (Web Text))

Max Activating Token Index: 1010

Text #3

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3680. Min Act: -0.0001

Data Index: 1525419 (Python Code)

Max Activating Token Index: 919

Text #4

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3619. Min Act: -0.0000

Data Index: 1468253 (Python Code)

Max Activating Token Index: 456

Text #5

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3485. Min Act: -0.0000

Data Index: 1429212 (Python Code)

Max Activating Token Index: 43

Text #6

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3626. Min Act: -0.0000

Data Index: 1442890 (Python Code)

Max Activating Token Index: 889

Text #7

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3818. Min Act: -0.0000

Data Index: 1434878 (Python Code)

Max Activating Token Index: 560

Text #8

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3431. Min Act: -0.0000

Data Index: 1523977 (Python Code)

Max Activating Token Index: 592

Text #9

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3325. Min Act: -0.0001

Data Index: 278708 (C4 (Web Text))

Max Activating Token Index: 840

Text #10

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3404. Min Act: -0.0000

Data Index: 1524042 (Python Code)

Max Activating Token Index: 192

Text #11

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3372. Min Act: -0.0001

Data Index: 1196273 (C4 (Web Text))

Max Activating Token Index: 323

Text #12

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3152. Min Act: -0.0001

Data Index: 1401473 (Python Code)

Max Activating Token Index: 527

Text #13

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3184. Min Act: -0.0001

Data Index: 111113 (C4 (Web Text))

Max Activating Token Index: 724

Text #14

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3436. Min Act: -0.0001

Data Index: 1619040 (Python Code)

Max Activating Token Index: 145

Text #15

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3540. Min Act: -0.0001

Data Index: 473488 (C4 (Web Text))

Max Activating Token Index: 837

Text #16

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3128. Min Act: -0.0001

Data Index: 1433818 (Python Code)

Max Activating Token Index: 508

Text #17

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3152. Min Act: -0.0001

Data Index: 321826 (C4 (Web Text))

Max Activating Token Index: 29

Text #18

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.3126. Min Act: -0.0001

Data Index: 455749 (C4 (Web Text))

Max Activating Token Index: 799

Text #19

Max Range: 0.5648. Min Range: -0.5648

Max Act: 0.2897. Min Act: -0.0000

Data Index: 1518615 (Python Code)

Max Activating Token Index: 764