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

    Neuronpedia

    INDEX
    Explanations

    distance

    np_max-act · gemini-2.0-flash

    The neuron responds to mentions of spatial distance, especially phrases indicating something is “far away.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     respects
    -0.08
    pending
    -0.08
    .linear
    -0.07
    iosa
    -0.07
     inspection
    -0.07
     combining
    -0.07
    Rose
    -0.07
     shadows
    -0.07
     Gulf
    -0.06
     collapse
    -0.06
    POSITIVE LOGITS
    컨
    0.08
    .keep
    0.07
    .Support
    0.07
    ụp
    0.06
    _qp
    0.06
    gew
    0.06
    :"-
    0.06
     Radius
    0.06
     مج
    0.06
     painstaking
    0.06
    Act Density 0.030%

    No Known Activations