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

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

Neuron 3008 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: 5.8530. Min Range: -5.8530

Max Act: 5.7922. Min Act: -0.0000

Data Index: 452533 (C4 (Web Text))

Max Activating Token Index: 1008

Text #1

Max Range: 5.8530. Min Range: -5.8530

Max Act: 5.8530. Min Act: -0.0000

Data Index: 361798 (C4 (Web Text))

Max Activating Token Index: 464

Text #2

Max Range: 5.8530. Min Range: -5.8530

Max Act: 5.6448. Min Act: -0.0000

Data Index: 877889 (C4 (Web Text))

Max Activating Token Index: 655

Text #3

Max Range: 5.8530. Min Range: -5.8530

Max Act: 5.1726. Min Act: -0.0000

Data Index: 1118279 (C4 (Web Text))

Max Activating Token Index: 434

Text #4

Max Range: 5.8530. Min Range: -5.8530

Max Act: 5.3034. Min Act: -0.0000

Data Index: 325211 (C4 (Web Text))

Max Activating Token Index: 898

Text #5

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.9057. Min Act: -0.0000

Data Index: 1288577 (C4 (Web Text))

Max Activating Token Index: 913

Text #6

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.7931. Min Act: -0.0000

Data Index: 1591637 (Python Code)

Max Activating Token Index: 936

Text #7

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.5552. Min Act: -0.0000

Data Index: 748614 (C4 (Web Text))

Max Activating Token Index: 668

Text #8

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.3107. Min Act: -0.0000

Data Index: 799712 (C4 (Web Text))

Max Activating Token Index: 1001

Text #9

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.8251. Min Act: -0.0000

Data Index: 505731 (C4 (Web Text))

Max Activating Token Index: 743

Text #10

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.0422. Min Act: -0.0000

Data Index: 568366 (C4 (Web Text))

Max Activating Token Index: 372

Text #11

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.0262. Min Act: -0.0000

Data Index: 1070903 (C4 (Web Text))

Max Activating Token Index: 410

Text #12

Max Range: 5.8530. Min Range: -5.8530

Max Act: 3.9198. Min Act: -0.0000

Data Index: 737565 (C4 (Web Text))

Max Activating Token Index: 307

Text #13

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.2739. Min Act: -0.0000

Data Index: 844288 (C4 (Web Text))

Max Activating Token Index: 765

Text #14

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.0945. Min Act: -0.0000

Data Index: 1551413 (Python Code)

Max Activating Token Index: 344

Text #15

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.0787. Min Act: -0.0000

Data Index: 719983 (C4 (Web Text))

Max Activating Token Index: 197

Text #16

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.0670. Min Act: -0.0000

Data Index: 557722 (C4 (Web Text))

Max Activating Token Index: 214

Text #17

Max Range: 5.8530. Min Range: -5.8530

Max Act: 4.1874. Min Act: -0.0000

Data Index: 964155 (C4 (Web Text))

Max Activating Token Index: 201

Text #18

Max Range: 5.8530. Min Range: -5.8530

Max Act: 3.8893. Min Act: -0.0000

Data Index: 106470 (C4 (Web Text))

Max Activating Token Index: 887

Text #19

Max Range: 5.8530. Min Range: -5.8530

Max Act: 3.9423. Min Act: -0.0000

Data Index: 768136 (C4 (Web Text))

Max Activating Token Index: 542