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

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

Neuron 192 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

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

Text #0

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2745. Min Act: -0.0001

Data Index: 984802 (C4 (Web Text))

Max Activating Token Index: 952

Text #1

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2938. Min Act: -0.0001

Data Index: 727725 (C4 (Web Text))

Max Activating Token Index: 247

Text #2

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2636. Min Act: -0.0001

Data Index: 1011897 (C4 (Web Text))

Max Activating Token Index: 771

Text #3

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.3036. Min Act: -0.0001

Data Index: 723538 (C4 (Web Text))

Max Activating Token Index: 998

Text #4

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2718. Min Act: -0.0001

Data Index: 960676 (C4 (Web Text))

Max Activating Token Index: 991

Text #5

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2646. Min Act: -0.0001

Data Index: 542317 (C4 (Web Text))

Max Activating Token Index: 764

Text #6

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2836. Min Act: -0.0001

Data Index: 1094210 (C4 (Web Text))

Max Activating Token Index: 19

Text #7

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2891. Min Act: -0.0001

Data Index: 928286 (C4 (Web Text))

Max Activating Token Index: 851

Text #8

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2519. Min Act: -0.0001

Data Index: 873794 (C4 (Web Text))

Max Activating Token Index: 309

Text #9

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2711. Min Act: -0.0001

Data Index: 735181 (C4 (Web Text))

Max Activating Token Index: 790

Text #10

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2515. Min Act: -0.0001

Data Index: 1268229 (C4 (Web Text))

Max Activating Token Index: 770

Text #11

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2403. Min Act: -0.0001

Data Index: 1133760 (C4 (Web Text))

Max Activating Token Index: 587

Text #12

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2510. Min Act: -0.0001

Data Index: 796421 (C4 (Web Text))

Max Activating Token Index: 486

Text #13

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2494. Min Act: -0.0001

Data Index: 1221794 (C4 (Web Text))

Max Activating Token Index: 980

Text #14

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2490. Min Act: -0.0001

Data Index: 1092025 (C4 (Web Text))

Max Activating Token Index: 817

Text #15

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2541. Min Act: -0.0001

Data Index: 470888 (C4 (Web Text))

Max Activating Token Index: 680

Text #16

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2626. Min Act: -0.0001

Data Index: 243480 (C4 (Web Text))

Max Activating Token Index: 221

Text #17

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2529. Min Act: -0.0001

Data Index: 1038819 (C4 (Web Text))

Max Activating Token Index: 979

Text #18

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2345. Min Act: -0.0001

Data Index: 49630 (C4 (Web Text))

Max Activating Token Index: 268

Text #19

Max Range: 0.3036. Min Range: -0.3036

Max Act: 0.2332. Min Act: -0.0001

Data Index: 450874 (C4 (Web Text))

Max Activating Token Index: 1018