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

    Neuronpedia

    INDEX
    Explanations

    Immigration and citizenship

    np_max-act · gemini-2.0-flash

    The neuron responds to mentions of immigration policy proposals and related terms (e.g. “immigration,” “comprehensive,” “plan,” “path to citizenship,” “reform,” “legalization”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     disturbing
    -0.07
    _output
    -0.06
    Pooling
    -0.06
    ort
    -0.06
    (al
    -0.06
     Delete
    -0.06
    duction
    -0.06
    ian
    -0.06
    ovaného
    -0.06
    positive
    -0.06
    POSITIVE LOGITS
    _METADATA
    0.07
    ΟΥ
    0.07
     الى
    0.07
    dojo
    0.06
    ểm
    0.06
     hairstyle
    0.06
     brightness
    0.06
    _CHARACTER
    0.06
     výhod
    0.06
    tbl
    0.06
    Act Density 0.004%

    No Known Activations