Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    examples, then, returns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    diving
    1.29
     bowlers
    1.28
     quiere
    1.27
     Lipschitz
    1.16
     ازاي
    1.16
    fach
    1.16
     amazing
    1.15
    do
    1.15
     Gost
    1.14
     arched
    1.13
    POSITIVE LOGITS
     кола
    1.14
    ्रोल
    1.00
     crayons
    0.99
     pesky
    0.98
    েও
    0.98
    anish
    0.95
     dozen
    0.93
    నాలు
    0.93
     tablespoons
    0.92
    <td>
    0.90
    Act Density 0.000%

    No Known Activations