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

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

Neuron 3957 in Layer 10

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-12l')

Text #0

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.7876. Min Act: -0.0000

Data Index: 737890 (C4 (Web Text))

Max Activating Token Index: 247

Text #1

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.6816. Min Act: -0.0000

Data Index: 639191 (C4 (Web Text))

Max Activating Token Index: 450

Text #2

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.6723. Min Act: -0.0000

Data Index: 63519 (C4 (Web Text))

Max Activating Token Index: 965

Text #3

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.6320. Min Act: -0.0000

Data Index: 490808 (C4 (Web Text))

Max Activating Token Index: 550

Text #4

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.5560. Min Act: -0.0000

Data Index: 91984 (C4 (Web Text))

Max Activating Token Index: 272

Text #5

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.5269. Min Act: -0.0000

Data Index: 189146 (C4 (Web Text))

Max Activating Token Index: 869

Text #6

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4621. Min Act: -0.0000

Data Index: 146668 (C4 (Web Text))

Max Activating Token Index: 237

Text #7

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4686. Min Act: -0.0000

Data Index: 280311 (C4 (Web Text))

Max Activating Token Index: 680

Text #8

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4538. Min Act: -0.0000

Data Index: 1279716 (C4 (Web Text))

Max Activating Token Index: 571

Text #9

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4287. Min Act: -0.0000

Data Index: 1087859 (C4 (Web Text))

Max Activating Token Index: 783

Text #10

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4059. Min Act: -0.0000

Data Index: 1039750 (C4 (Web Text))

Max Activating Token Index: 191

Text #11

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4355. Min Act: -0.0000

Data Index: 1361380 (Python Code)

Max Activating Token Index: 214

Text #12

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4207. Min Act: -0.0000

Data Index: 1452607 (Python Code)

Max Activating Token Index: 451

Text #13

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4516. Min Act: -0.0000

Data Index: 1371189 (Python Code)

Max Activating Token Index: 425

Text #14

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4105. Min Act: -0.0000

Data Index: 363826 (C4 (Web Text))

Max Activating Token Index: 609

Text #15

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4041. Min Act: 0.0000

Data Index: 1615972 (Python Code)

Max Activating Token Index: 541

Text #16

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4001. Min Act: -0.0000

Data Index: 1382720 (Python Code)

Max Activating Token Index: 436

Text #17

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.3968. Min Act: -0.0000

Data Index: 1532319 (Python Code)

Max Activating Token Index: 996

Text #18

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.4063. Min Act: -0.0000

Data Index: 723615 (C4 (Web Text))

Max Activating Token Index: 326

Text #19

Max Range: 0.7876. Min Range: -0.7876

Max Act: 0.3873. Min Act: -0.0000

Data Index: 1645547 (Python Code)

Max Activating Token Index: 505