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

    Neuronpedia

    INDEX
    Explanations

    set followed by a word

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Part
    0.76
    疮
    0.73
     spirals
    0.64
     Spiral
    0.62
     spiral
    0.61
    sted
    0.60
    τικο
    0.60
    मिता
    0.59
    spiral
    0.59
    ebra
    0.59
    POSITIVE LOGITS
     chargés
    0.82
    ﺌ
    0.81
    וע
    0.80
    肪
    0.80
     Goldsmith
    0.79
     গবে
    0.78
    raron
    0.78
    祺
    0.78
    gani
    0.77
     blames
    0.77
    Act Density 0.021%

    No Known Activations