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

    Neuronpedia

    INDEX
    Explanations

    numerical calculations

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on numeric tokens (i.e. individual numbers or digit sequences).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    렇게
    -0.07
     SV
    -0.07
     marble
    -0.06
     TH
    -0.06
     sagte
    -0.06
    Channel
    -0.06
    _FAST
    -0.06
     hud
    -0.06
    JUST
    -0.06
    _to
    -0.06
    POSITIVE LOGITS
    готов
    0.07
     التش
    0.07
     Autom
    0.06
     đoàn
    0.06
     Titanium
    0.06
     incap
    0.06
     розвитку
    0.06
    ursed
    0.06
     хроничес
    0.06
    mA
    0.06
    Act Density 0.011%

    No Known Activations