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

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

Neuron 4768 in Layer 8

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 3.1652. Min Range: -3.1652

Max Act: 3.1652. Min Act: -0.0000

Data Index: 756104 (C4 (Web Text))

Max Activating Token Index: 431

Text #1

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.8027. Min Act: -0.0000

Data Index: 263826 (C4 (Web Text))

Max Activating Token Index: 89

Text #2

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.4526. Min Act: -0.0000

Data Index: 912472 (C4 (Web Text))

Max Activating Token Index: 618

Text #3

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.2331. Min Act: -0.0000

Data Index: 1131939 (C4 (Web Text))

Max Activating Token Index: 336

Text #4

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.3202. Min Act: -0.0000

Data Index: 607336 (C4 (Web Text))

Max Activating Token Index: 79

Text #5

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.1703. Min Act: -0.0000

Data Index: 222208 (C4 (Web Text))

Max Activating Token Index: 932

Text #6

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.2977. Min Act: -0.0000

Data Index: 1115750 (C4 (Web Text))

Max Activating Token Index: 907

Text #7

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.2368. Min Act: -0.0000

Data Index: 258703 (C4 (Web Text))

Max Activating Token Index: 789

Text #8

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.2452. Min Act: -0.0000

Data Index: 898165 (C4 (Web Text))

Max Activating Token Index: 430

Text #9

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0924. Min Act: -0.0000

Data Index: 1275759 (C4 (Web Text))

Max Activating Token Index: 460

Text #10

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0553. Min Act: -0.0000

Data Index: 690756 (C4 (Web Text))

Max Activating Token Index: 543

Text #11

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0539. Min Act: -0.0000

Data Index: 424583 (C4 (Web Text))

Max Activating Token Index: 869

Text #12

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0552. Min Act: -0.0000

Data Index: 151642 (C4 (Web Text))

Max Activating Token Index: 587

Text #13

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0403. Min Act: -0.0000

Data Index: 668175 (C4 (Web Text))

Max Activating Token Index: 440

Text #14

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.1188. Min Act: -0.0000

Data Index: 347724 (C4 (Web Text))

Max Activating Token Index: 685

Text #15

Max Range: 3.1652. Min Range: -3.1652

Max Act: 2.0873. Min Act: -0.0000

Data Index: 540186 (C4 (Web Text))

Max Activating Token Index: 135

Text #16

Max Range: 3.1652. Min Range: -3.1652

Max Act: 1.9996. Min Act: -0.0000

Data Index: 941214 (C4 (Web Text))

Max Activating Token Index: 774

Text #17

Max Range: 3.1652. Min Range: -3.1652

Max Act: 1.8855. Min Act: -0.0000

Data Index: 18015 (C4 (Web Text))

Max Activating Token Index: 621

Text #18

Max Range: 3.1652. Min Range: -3.1652

Max Act: 1.9454. Min Act: -0.0000

Data Index: 766444 (C4 (Web Text))

Max Activating Token Index: 854

Text #19

Max Range: 3.1652. Min Range: -3.1652

Max Act: 1.9253. Min Act: -0.0000

Data Index: 1108033 (C4 (Web Text))

Max Activating Token Index: 762