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

    Neuronpedia

    INDEX
    Explanations

    open source weights

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    related
    1.15
    based
    1.13
    type
    1.03
    এর
    1.01
    driven
    0.99
    like
    0.95
    style
    0.94
    t
    0.92
    G
    0.91
    induced
    0.90
    POSITIVE LOGITS
    -
    1.27
     unnamed
    0.94
    ـ
    0.93
     elem
    0.93
    ؘ
    0.91
     wɔ
    0.91
     deze
    0.89
    ـــ
    0.89
     isomeric
    0.88
     chem
    0.88
    Act Density 0.121%

    No Known Activations