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

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

Neuron 1366 in Layer 9

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.7826. Min Act: -0.0000

Data Index: 1369451 (Python Code)

Max Activating Token Index: 63

Text #1

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.7353. Min Act: -0.0000

Data Index: 407155 (C4 (Web Text))

Max Activating Token Index: 243

Text #2

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.7566. Min Act: -0.0000

Data Index: 1648912 (Python Code)

Max Activating Token Index: 116

Text #3

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.6582. Min Act: -0.0000

Data Index: 402790 (C4 (Web Text))

Max Activating Token Index: 901

Text #4

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.6802. Min Act: -0.0000

Data Index: 565232 (C4 (Web Text))

Max Activating Token Index: 303

Text #5

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.6316. Min Act: -0.0000

Data Index: 459179 (C4 (Web Text))

Max Activating Token Index: 763

Text #6

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.6108. Min Act: -0.0000

Data Index: 1604062 (Python Code)

Max Activating Token Index: 730

Text #7

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5409. Min Act: -0.0000

Data Index: 357876 (C4 (Web Text))

Max Activating Token Index: 208

Text #8

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5950. Min Act: -0.0000

Data Index: 1377077 (Python Code)

Max Activating Token Index: 155

Text #9

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5732. Min Act: -0.0000

Data Index: 164064 (C4 (Web Text))

Max Activating Token Index: 724

Text #10

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5466. Min Act: -0.0000

Data Index: 791972 (C4 (Web Text))

Max Activating Token Index: 849

Text #11

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5673. Min Act: -0.0000

Data Index: 345903 (C4 (Web Text))

Max Activating Token Index: 670

Text #12

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5228. Min Act: -0.0000

Data Index: 757213 (C4 (Web Text))

Max Activating Token Index: 684

Text #13

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5670. Min Act: -0.0000

Data Index: 1604302 (Python Code)

Max Activating Token Index: 572

Text #14

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5165. Min Act: -0.0000

Data Index: 682880 (C4 (Web Text))

Max Activating Token Index: 471

Text #15

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5184. Min Act: -0.0000

Data Index: 831848 (C4 (Web Text))

Max Activating Token Index: 834

Text #16

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5672. Min Act: -0.0000

Data Index: 1571057 (Python Code)

Max Activating Token Index: 274

Text #17

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.5104. Min Act: -0.0000

Data Index: 1493660 (Python Code)

Max Activating Token Index: 490

Text #18

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.4969. Min Act: -0.0000

Data Index: 1617466 (Python Code)

Max Activating Token Index: 473

Text #19

Max Range: 0.7826. Min Range: -0.7826

Max Act: 0.4797. Min Act: -0.0000

Data Index: 605566 (C4 (Web Text))

Max Activating Token Index: 237