Model: SoLU Model: 4 Layers, 2048 Neurons per Layer

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

Neuron 590 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-4l')

Text #0

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0725. Min Act: -0.0002

Data Index: 1498173 (Python Code)

Max Activating Token Index: 789

Text #1

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0775. Min Act: -0.0002

Data Index: 1413070 (Python Code)

Max Activating Token Index: 302

Text #2

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0799. Min Act: -0.0002

Data Index: 1295927 (C4 (Web Text))

Max Activating Token Index: 606

Text #3

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0743. Min Act: -0.0002

Data Index: 1445628 (Python Code)

Max Activating Token Index: 121

Text #4

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0732. Min Act: -0.0002

Data Index: 121968 (C4 (Web Text))

Max Activating Token Index: 492

Text #5

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0693. Min Act: -0.0002

Data Index: 1153004 (C4 (Web Text))

Max Activating Token Index: 385

Text #6

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0697. Min Act: -0.0002

Data Index: 957277 (C4 (Web Text))

Max Activating Token Index: 1013

Text #7

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0715. Min Act: -0.0002

Data Index: 1475372 (Python Code)

Max Activating Token Index: 400

Text #8

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0594. Min Act: -0.0002

Data Index: 1197297 (C4 (Web Text))

Max Activating Token Index: 709

Text #9

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0669. Min Act: -0.0002

Data Index: 1486693 (Python Code)

Max Activating Token Index: 23

Text #10

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0735. Min Act: -0.0002

Data Index: 1304038 (C4 (Web Text))

Max Activating Token Index: 899

Text #11

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0685. Min Act: -0.0002

Data Index: 1398871 (Python Code)

Max Activating Token Index: 141

Text #12

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0691. Min Act: -0.0002

Data Index: 11430 (C4 (Web Text))

Max Activating Token Index: 451

Text #13

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0683. Min Act: -0.0002

Data Index: 1431655 (Python Code)

Max Activating Token Index: 206

Text #14

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0666. Min Act: -0.0002

Data Index: 715593 (C4 (Web Text))

Max Activating Token Index: 605

Text #15

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0669. Min Act: -0.0002

Data Index: 1374360 (Python Code)

Max Activating Token Index: 136

Text #16

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0638. Min Act: -0.0002

Data Index: 321985 (C4 (Web Text))

Max Activating Token Index: 640

Text #17

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0690. Min Act: -0.0002

Data Index: 584183 (C4 (Web Text))

Max Activating Token Index: 601

Text #18

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0623. Min Act: -0.0002

Data Index: 1530446 (Python Code)

Max Activating Token Index: 471

Text #19

Max Range: 0.0799. Min Range: -0.0799

Max Act: 0.0662. Min Act: -0.0002

Data Index: 1393552 (Python Code)

Max Activating Token Index: 488