Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

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

Neuron 346 in Layer 3

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2903. Min Act: -0.0001

Data Index: 2876 (C4 (Web Text))

Max Activating Token Index: 492

Text #1

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.3059. Min Act: -0.0001

Data Index: 385230 (C4 (Web Text))

Max Activating Token Index: 179

Text #2

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2968. Min Act: -0.0001

Data Index: 1103650 (C4 (Web Text))

Max Activating Token Index: 276

Text #3

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.3011. Min Act: -0.0001

Data Index: 89365 (C4 (Web Text))

Max Activating Token Index: 90

Text #4

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2988. Min Act: -0.0001

Data Index: 882937 (C4 (Web Text))

Max Activating Token Index: 349

Text #5

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.3059. Min Act: -0.0001

Data Index: 464364 (C4 (Web Text))

Max Activating Token Index: 63

Text #6

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2845. Min Act: -0.0001

Data Index: 140852 (C4 (Web Text))

Max Activating Token Index: 14

Text #7

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2927. Min Act: -0.0001

Data Index: 1211083 (C4 (Web Text))

Max Activating Token Index: 36

Text #8

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2759. Min Act: -0.0001

Data Index: 627336 (C4 (Web Text))

Max Activating Token Index: 288

Text #9

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2794. Min Act: -0.0001

Data Index: 241116 (C4 (Web Text))

Max Activating Token Index: 529

Text #10

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2796. Min Act: -0.0001

Data Index: 1302114 (C4 (Web Text))

Max Activating Token Index: 841

Text #11

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2666. Min Act: -0.0001

Data Index: 794985 (C4 (Web Text))

Max Activating Token Index: 40

Text #12

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2561. Min Act: -0.0001

Data Index: 544011 (C4 (Web Text))

Max Activating Token Index: 535

Text #13

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2463. Min Act: -0.0001

Data Index: 1133908 (C4 (Web Text))

Max Activating Token Index: 936

Text #14

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2610. Min Act: -0.0001

Data Index: 1223943 (C4 (Web Text))

Max Activating Token Index: 78

Text #15

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2473. Min Act: -0.0001

Data Index: 254822 (C4 (Web Text))

Max Activating Token Index: 119

Text #16

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2638. Min Act: -0.0001

Data Index: 1251911 (C4 (Web Text))

Max Activating Token Index: 591

Text #17

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2527. Min Act: -0.0001

Data Index: 344967 (C4 (Web Text))

Max Activating Token Index: 730

Text #18

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2532. Min Act: -0.0001

Data Index: 548382 (C4 (Web Text))

Max Activating Token Index: 295

Text #19

Max Range: 0.3059. Min Range: -0.3059

Max Act: 0.2456. Min Act: -0.0001

Data Index: 943785 (C4 (Web Text))

Max Activating Token Index: 885