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

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

Neuron 5681 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.8672. Min Act: -0.0000

Data Index: 1621628 (Python Code)

Max Activating Token Index: 123

Text #1

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7722. Min Act: -0.0000

Data Index: 778373 (C4 (Web Text))

Max Activating Token Index: 680

Text #2

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.8338. Min Act: -0.0000

Data Index: 946741 (C4 (Web Text))

Max Activating Token Index: 1020

Text #3

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.8174. Min Act: -0.0000

Data Index: 937618 (C4 (Web Text))

Max Activating Token Index: 119

Text #4

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.8059. Min Act: -0.0000

Data Index: 172721 (C4 (Web Text))

Max Activating Token Index: 205

Text #5

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.8214. Min Act: -0.0000

Data Index: 1635597 (Python Code)

Max Activating Token Index: 443

Text #6

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7442. Min Act: -0.0000

Data Index: 678976 (C4 (Web Text))

Max Activating Token Index: 307

Text #7

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7855. Min Act: -0.0000

Data Index: 634874 (C4 (Web Text))

Max Activating Token Index: 1021

Text #8

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7319. Min Act: -0.0000

Data Index: 281010 (C4 (Web Text))

Max Activating Token Index: 802

Text #9

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6963. Min Act: -0.0000

Data Index: 1050386 (C4 (Web Text))

Max Activating Token Index: 593

Text #10

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7131. Min Act: -0.0000

Data Index: 153626 (C4 (Web Text))

Max Activating Token Index: 425

Text #11

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6737. Min Act: -0.0000

Data Index: 58911 (C4 (Web Text))

Max Activating Token Index: 780

Text #12

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.7369. Min Act: -0.0000

Data Index: 1547889 (Python Code)

Max Activating Token Index: 72

Text #13

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6531. Min Act: -0.0000

Data Index: 935343 (C4 (Web Text))

Max Activating Token Index: 413

Text #14

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6971. Min Act: -0.0000

Data Index: 497387 (C4 (Web Text))

Max Activating Token Index: 451

Text #15

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6317. Min Act: -0.0000

Data Index: 1476789 (Python Code)

Max Activating Token Index: 616

Text #16

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6447. Min Act: -0.0000

Data Index: 1129449 (C4 (Web Text))

Max Activating Token Index: 577

Text #17

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6582. Min Act: -0.0000

Data Index: 1550573 (Python Code)

Max Activating Token Index: 913

Text #18

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6300. Min Act: -0.0000

Data Index: 365032 (C4 (Web Text))

Max Activating Token Index: 759

Text #19

Max Range: 0.8672. Min Range: -0.8672

Max Act: 0.6022. Min Act: -0.0000

Data Index: 233309 (C4 (Web Text))

Max Activating Token Index: 817