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

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

Neuron 3769 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.9099. Min Act: -0.0000

Data Index: 617053 (C4 (Web Text))

Max Activating Token Index: 482

Text #1

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8648. Min Act: -0.0000

Data Index: 230594 (C4 (Web Text))

Max Activating Token Index: 598

Text #2

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8762. Min Act: -0.0000

Data Index: 670111 (C4 (Web Text))

Max Activating Token Index: 103

Text #3

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8445. Min Act: -0.0000

Data Index: 347276 (C4 (Web Text))

Max Activating Token Index: 579

Text #4

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8159. Min Act: -0.0000

Data Index: 993752 (C4 (Web Text))

Max Activating Token Index: 622

Text #5

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8563. Min Act: -0.0000

Data Index: 1622302 (Python Code)

Max Activating Token Index: 507

Text #6

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7613. Min Act: -0.0000

Data Index: 1595866 (Python Code)

Max Activating Token Index: 77

Text #7

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8266. Min Act: -0.0000

Data Index: 1316842 (C4 (Web Text))

Max Activating Token Index: 567

Text #8

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8012. Min Act: -0.0000

Data Index: 564560 (C4 (Web Text))

Max Activating Token Index: 46

Text #9

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.8484. Min Act: -0.0000

Data Index: 756897 (C4 (Web Text))

Max Activating Token Index: 34

Text #10

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7641. Min Act: -0.0000

Data Index: 205282 (C4 (Web Text))

Max Activating Token Index: 537

Text #11

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7793. Min Act: -0.0000

Data Index: 1252575 (C4 (Web Text))

Max Activating Token Index: 251

Text #12

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7534. Min Act: -0.0000

Data Index: 369458 (C4 (Web Text))

Max Activating Token Index: 882

Text #13

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7668. Min Act: -0.0000

Data Index: 825551 (C4 (Web Text))

Max Activating Token Index: 394

Text #14

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7569. Min Act: -0.0000

Data Index: 1062174 (C4 (Web Text))

Max Activating Token Index: 437

Text #15

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7763. Min Act: -0.0000

Data Index: 1546146 (Python Code)

Max Activating Token Index: 121

Text #16

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7508. Min Act: -0.0000

Data Index: 911701 (C4 (Web Text))

Max Activating Token Index: 524

Text #17

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7530. Min Act: -0.0000

Data Index: 890476 (C4 (Web Text))

Max Activating Token Index: 277

Text #18

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7230. Min Act: -0.0000

Data Index: 612647 (C4 (Web Text))

Max Activating Token Index: 515

Text #19

Max Range: 0.9099. Min Range: -0.9099

Max Act: 0.7185. Min Act: -0.0000

Data Index: 1149201 (C4 (Web Text))

Max Activating Token Index: 180