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

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

Neuron 641 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: 0.3099. Min Range: -0.3099

Max Act: 0.2922. Min Act: -0.0000

Data Index: 1042489 (C4 (Web Text))

Max Activating Token Index: 603

Text #1

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.3099. Min Act: -0.0000

Data Index: 649139 (C4 (Web Text))

Max Activating Token Index: 999

Text #2

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2610. Min Act: -0.0000

Data Index: 1437188 (Python Code)

Max Activating Token Index: 633

Text #3

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2452. Min Act: -0.0000

Data Index: 101161 (C4 (Web Text))

Max Activating Token Index: 932

Text #4

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2200. Min Act: -0.0000

Data Index: 1640266 (Python Code)

Max Activating Token Index: 841

Text #5

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2299. Min Act: -0.0000

Data Index: 882900 (C4 (Web Text))

Max Activating Token Index: 74

Text #6

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2526. Min Act: -0.0000

Data Index: 1149989 (C4 (Web Text))

Max Activating Token Index: 596

Text #7

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2129. Min Act: -0.0000

Data Index: 1264004 (C4 (Web Text))

Max Activating Token Index: 181

Text #8

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2042. Min Act: -0.0000

Data Index: 741807 (C4 (Web Text))

Max Activating Token Index: 918

Text #9

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2018. Min Act: -0.0000

Data Index: 680102 (C4 (Web Text))

Max Activating Token Index: 759

Text #10

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1907. Min Act: -0.0000

Data Index: 824050 (C4 (Web Text))

Max Activating Token Index: 294

Text #11

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1976. Min Act: -0.0000

Data Index: 4199 (C4 (Web Text))

Max Activating Token Index: 328

Text #12

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.2016. Min Act: -0.0000

Data Index: 600574 (C4 (Web Text))

Max Activating Token Index: 193

Text #13

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1992. Min Act: -0.0000

Data Index: 1199983 (C4 (Web Text))

Max Activating Token Index: 284

Text #14

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1898. Min Act: -0.0000

Data Index: 704155 (C4 (Web Text))

Max Activating Token Index: 593

Text #15

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1857. Min Act: -0.0000

Data Index: 760464 (C4 (Web Text))

Max Activating Token Index: 279

Text #16

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1942. Min Act: -0.0000

Data Index: 1489238 (Python Code)

Max Activating Token Index: 741

Text #17

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1855. Min Act: -0.0000

Data Index: 505042 (C4 (Web Text))

Max Activating Token Index: 331

Text #18

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1789. Min Act: -0.0000

Data Index: 182401 (C4 (Web Text))

Max Activating Token Index: 591

Text #19

Max Range: 0.3099. Min Range: -0.3099

Max Act: 0.1837. Min Act: -0.0000

Data Index: 529962 (C4 (Web Text))

Max Activating Token Index: 584