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

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

Neuron 630 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3892. Min Act: -0.0000

Data Index: 972524 (C4 (Web Text))

Max Activating Token Index: 1010

Text #1

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3713. Min Act: -0.0000

Data Index: 90419 (C4 (Web Text))

Max Activating Token Index: 747

Text #2

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3568. Min Act: -0.0001

Data Index: 1503879 (Python Code)

Max Activating Token Index: 92

Text #3

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3330. Min Act: -0.0000

Data Index: 189083 (C4 (Web Text))

Max Activating Token Index: 70

Text #4

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3386. Min Act: -0.0000

Data Index: 463527 (C4 (Web Text))

Max Activating Token Index: 747

Text #5

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3184. Min Act: -0.0000

Data Index: 314886 (C4 (Web Text))

Max Activating Token Index: 794

Text #6

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3022. Min Act: -0.0001

Data Index: 696013 (C4 (Web Text))

Max Activating Token Index: 1018

Text #7

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3215. Min Act: -0.0000

Data Index: 769267 (C4 (Web Text))

Max Activating Token Index: 496

Text #8

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3033. Min Act: -0.0001

Data Index: 303271 (C4 (Web Text))

Max Activating Token Index: 43

Text #9

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2963. Min Act: -0.0001

Data Index: 827332 (C4 (Web Text))

Max Activating Token Index: 200

Text #10

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2933. Min Act: -0.0000

Data Index: 1303882 (C4 (Web Text))

Max Activating Token Index: 503

Text #11

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.3030. Min Act: -0.0001

Data Index: 919356 (C4 (Web Text))

Max Activating Token Index: 281

Text #12

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2885. Min Act: -0.0000

Data Index: 687115 (C4 (Web Text))

Max Activating Token Index: 994

Text #13

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2975. Min Act: -0.0000

Data Index: 183539 (C4 (Web Text))

Max Activating Token Index: 45

Text #14

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2844. Min Act: -0.0000

Data Index: 811448 (C4 (Web Text))

Max Activating Token Index: 256

Text #15

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2661. Min Act: -0.0001

Data Index: 1525294 (Python Code)

Max Activating Token Index: 959

Text #16

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2706. Min Act: -0.0000

Data Index: 457733 (C4 (Web Text))

Max Activating Token Index: 673

Text #17

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2674. Min Act: -0.0000

Data Index: 1018989 (C4 (Web Text))

Max Activating Token Index: 810

Text #18

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2758. Min Act: -0.0001

Data Index: 324625 (C4 (Web Text))

Max Activating Token Index: 269

Text #19

Max Range: 0.3892. Min Range: -0.3892

Max Act: 0.2732. Min Act: -0.0001

Data Index: 973368 (C4 (Web Text))

Max Activating Token Index: 738