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

    Neuronpedia

    INDEX
    Explanations

    salt, sweet, and syrupy ingredients

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fall
    1.61
     secondary
    1.60
     dugg
    1.60
    ies
    1.52
     thousands
    1.47
     E
    1.46
     excavated
    1.43
     slump
    1.43
     cak
    1.43
     ’
    1.41
    POSITIVE LOGITS
    ו
    3.03
    iť
    2.75
    ه
    2.64
    ומי
    2.64
    र
    2.43
    ený
    2.40
    ה
    2.40
    kker
    2.36
    י
    2.36
    glise
    2.36
    Act Density 0.130%

    No Known Activations