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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    .SubElement
    -0.07
    deaux
    -0.06
    เง
    -0.06
    .nextSibling
    -0.06
     hayal
    -0.06
    .Unsupported
    -0.06
     devuelve
    -0.06
     chees
    -0.06
    Replacing
    -0.06
     �
    -0.06
    POSITIVE LOGITS
    (term
    0.07
    —we
    0.06
     Т
    0.06
    son
    0.06
    (files
    0.06
    сон
    0.06
    _bind
    0.06
    änger
    0.06
    ising
    0.06
     neighbourhood
    0.06
    Act Density 0.144%

    No Known Activations