Model: SoLU Model: 8 Layers, 4096 Neurons per Layer

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

Neuron 3502 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-8l')

Text #0

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.5702. Min Act: -0.0001

Data Index: 540144 (C4 (Web Text))

Max Activating Token Index: 101

Text #1

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.4959. Min Act: -0.0001

Data Index: 264609 (C4 (Web Text))

Max Activating Token Index: 444

Text #2

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.5193. Min Act: -0.0001

Data Index: 1635981 (Python Code)

Max Activating Token Index: 552

Text #3

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.3432. Min Act: -0.0001

Data Index: 1248348 (C4 (Web Text))

Max Activating Token Index: 568

Text #4

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.2827. Min Act: -0.0001

Data Index: 848912 (C4 (Web Text))

Max Activating Token Index: 264

Text #5

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.3235. Min Act: -0.0001

Data Index: 857032 (C4 (Web Text))

Max Activating Token Index: 969

Text #6

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.2405. Min Act: -0.0001

Data Index: 1604210 (Python Code)

Max Activating Token Index: 123

Text #7

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.1121. Min Act: -0.0001

Data Index: 437542 (C4 (Web Text))

Max Activating Token Index: 154

Text #8

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.1649. Min Act: -0.0001

Data Index: 591764 (C4 (Web Text))

Max Activating Token Index: 218

Text #9

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.1213. Min Act: -0.0001

Data Index: 325433 (C4 (Web Text))

Max Activating Token Index: 235

Text #10

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0197. Min Act: -0.0001

Data Index: 1430376 (Python Code)

Max Activating Token Index: 865

Text #11

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0666. Min Act: -0.0001

Data Index: 916408 (C4 (Web Text))

Max Activating Token Index: 845

Text #12

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0845. Min Act: -0.0001

Data Index: 856014 (C4 (Web Text))

Max Activating Token Index: 347

Text #13

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0262. Min Act: -0.0001

Data Index: 1332713 (C4 (Web Text))

Max Activating Token Index: 950

Text #14

Max Range: 1.5702. Min Range: -1.5702

Max Act: 0.9519. Min Act: -0.0001

Data Index: 275840 (C4 (Web Text))

Max Activating Token Index: 414

Text #15

Max Range: 1.5702. Min Range: -1.5702

Max Act: 0.9915. Min Act: -0.0001

Data Index: 463544 (C4 (Web Text))

Max Activating Token Index: 618

Text #16

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0032. Min Act: -0.0001

Data Index: 1312890 (C4 (Web Text))

Max Activating Token Index: 628

Text #17

Max Range: 1.5702. Min Range: -1.5702

Max Act: 1.0061. Min Act: -0.0001

Data Index: 1212978 (C4 (Web Text))

Max Activating Token Index: 616

Text #18

Max Range: 1.5702. Min Range: -1.5702

Max Act: 0.9924. Min Act: -0.0001

Data Index: 29449 (C4 (Web Text))

Max Activating Token Index: 124

Text #19

Max Range: 1.5702. Min Range: -1.5702

Max Act: 0.9679. Min Act: -0.0001

Data Index: 840773 (C4 (Web Text))

Max Activating Token Index: 588