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

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

Neuron 3904 in Layer 6

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7136. Min Act: -0.0000

Data Index: 1483783 (Python Code)

Max Activating Token Index: 173

Text #1

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7889. Min Act: -0.0000

Data Index: 1623101 (Python Code)

Max Activating Token Index: 182

Text #2

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7778. Min Act: -0.0000

Data Index: 1508106 (Python Code)

Max Activating Token Index: 241

Text #3

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7152. Min Act: -0.0000

Data Index: 913025 (C4 (Web Text))

Max Activating Token Index: 700

Text #4

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7393. Min Act: -0.0000

Data Index: 1508961 (Python Code)

Max Activating Token Index: 77

Text #5

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6555. Min Act: -0.0000

Data Index: 1631888 (Python Code)

Max Activating Token Index: 281

Text #6

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6557. Min Act: -0.0000

Data Index: 1513689 (Python Code)

Max Activating Token Index: 363

Text #7

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6861. Min Act: -0.0000

Data Index: 1494993 (Python Code)

Max Activating Token Index: 359

Text #8

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6660. Min Act: -0.0000

Data Index: 1588105 (Python Code)

Max Activating Token Index: 66

Text #9

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.7005. Min Act: -0.0000

Data Index: 880727 (C4 (Web Text))

Max Activating Token Index: 875

Text #10

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.5834. Min Act: -0.0000

Data Index: 1648516 (Python Code)

Max Activating Token Index: 313

Text #11

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6445. Min Act: -0.0000

Data Index: 1399872 (Python Code)

Max Activating Token Index: 136

Text #12

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6805. Min Act: -0.0000

Data Index: 964857 (C4 (Web Text))

Max Activating Token Index: 919

Text #13

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6267. Min Act: -0.0000

Data Index: 1361440 (Python Code)

Max Activating Token Index: 157

Text #14

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6860. Min Act: -0.0000

Data Index: 1395052 (Python Code)

Max Activating Token Index: 235

Text #15

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6054. Min Act: -0.0000

Data Index: 57602 (C4 (Web Text))

Max Activating Token Index: 719

Text #16

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.6376. Min Act: -0.0000

Data Index: 289543 (C4 (Web Text))

Max Activating Token Index: 905

Text #17

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.5912. Min Act: -0.0000

Data Index: 1554430 (Python Code)

Max Activating Token Index: 141

Text #18

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.5732. Min Act: -0.0000

Data Index: 1390182 (Python Code)

Max Activating Token Index: 328

Text #19

Max Range: 0.7889. Min Range: -0.7889

Max Act: 0.5932. Min Act: -0.0000

Data Index: 1401460 (Python Code)

Max Activating Token Index: 394