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

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

Neuron 760 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 4.1703. Min Range: -4.1703

Max Act: 4.1703. Min Act: -0.0000

Data Index: 1613162 (Python Code)

Max Activating Token Index: 816

Text #1

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.9702. Min Act: -0.0000

Data Index: 741590 (C4 (Web Text))

Max Activating Token Index: 68

Text #2

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.9562. Min Act: -0.0000

Data Index: 1370942 (Python Code)

Max Activating Token Index: 15

Text #3

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.7977. Min Act: -0.0000

Data Index: 1117033 (C4 (Web Text))

Max Activating Token Index: 607

Text #4

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.6801. Min Act: -0.0000

Data Index: 1363499 (Python Code)

Max Activating Token Index: 476

Text #5

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4788. Min Act: -0.0000

Data Index: 283276 (C4 (Web Text))

Max Activating Token Index: 336

Text #6

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.5907. Min Act: -0.0000

Data Index: 1520370 (Python Code)

Max Activating Token Index: 683

Text #7

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.5882. Min Act: -0.0000

Data Index: 1415696 (Python Code)

Max Activating Token Index: 521

Text #8

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.5374. Min Act: -0.0000

Data Index: 158565 (C4 (Web Text))

Max Activating Token Index: 806

Text #9

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4392. Min Act: -0.0000

Data Index: 1022600 (C4 (Web Text))

Max Activating Token Index: 636

Text #10

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4114. Min Act: -0.0000

Data Index: 1487901 (Python Code)

Max Activating Token Index: 535

Text #11

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4132. Min Act: -0.0000

Data Index: 1576197 (Python Code)

Max Activating Token Index: 696

Text #12

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.3293. Min Act: -0.0000

Data Index: 1375373 (Python Code)

Max Activating Token Index: 634

Text #13

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4771. Min Act: -0.0000

Data Index: 1510875 (Python Code)

Max Activating Token Index: 77

Text #14

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.4615. Min Act: -0.0000

Data Index: 1432855 (Python Code)

Max Activating Token Index: 348

Text #15

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.3228. Min Act: -0.0000

Data Index: 442868 (C4 (Web Text))

Max Activating Token Index: 720

Text #16

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.1284. Min Act: -0.0000

Data Index: 324516 (C4 (Web Text))

Max Activating Token Index: 375

Text #17

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.3060. Min Act: -0.0000

Data Index: 1493592 (Python Code)

Max Activating Token Index: 653

Text #18

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.2672. Min Act: -0.0000

Data Index: 1480691 (Python Code)

Max Activating Token Index: 648

Text #19

Max Range: 4.1703. Min Range: -4.1703

Max Act: 3.1111. Min Act: -0.0000

Data Index: 1396870 (Python Code)

Max Activating Token Index: 958