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

    Neuronpedia

    INDEX
    Explanations

    Missing or undefined

    np_max-act · gemini-2.0-flash

    This neuron detects phrases indicating a default-when-unspecified clause (e.g. “si no se especifica nada”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     PC
    -0.06
     regulating
    -0.06
    ños
    -0.06
     ridiculous
    -0.06
    Try
    -0.06
    ECH
    -0.06
     pc
    -0.06
     grate
    -0.06
    him
    -0.06
     Try
    -0.06
    POSITIVE LOGITS
    .Pull
    0.07
    raise
    0.07
    庄
    0.06
    &q
    0.06
    -che
    0.06
     губер
    0.06
    516
    0.06
    ="//
    0.06
    &type
    0.06
    pell
    0.06
    Act Density 0.013%

    No Known Activations