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

    Neuronpedia

    INDEX
    Explanations

    Unusual text fragments

    np_max-act · gemini-2.0-flash

    This neuron activates on short code‐style identifiers (e.g. one‐ or two‐letter function or variable names) in programming text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    πει
    -0.08
     만
    -0.07
    ufac
    -0.07
     стоя
    -0.07
     đặc
    -0.06
     فروش
    -0.06
    alloc
    -0.06
    (interface
    -0.06
    Of
    -0.06
    wechat
    -0.06
    POSITIVE LOGITS
    .setOnClickListener
    0.06
     compassionate
    0.06
    rina
    0.06
    ，她
    0.06
     aşam
    0.06
    //----------------------------------------------------------------------------------------------------------------
    0.06
    _BR
    0.06
     cones
    0.06
    profil
    0.06
    .predicate
    0.06
    Act Density 7.871%

    No Known Activations