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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron activates on tokens that are part of date expressions (e.g. month names, day numbers, and the word “On” in temporal phrases).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     formulario
    -0.07
    ороз
    -0.07
    iaux
    -0.07
    kie
    -0.07
     ambassador
    -0.06
    .parseDouble
    -0.06
     illegally
    -0.06
    -suite
    -0.06
    've
    -0.06
     Server
    -0.06
    POSITIVE LOGITS
     Accom
    0.06
    _git
    0.06
     마음
    0.06
     động
    0.06
     IDS
    0.06
     Natur
    0.06
    .MESSAGE
    0.06
    üre
    0.06
     назнач
    0.06
    .safe
    0.06
    Act Density 0.003%

    No Known Activations