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

    Neuronpedia

    INDEX
    Explanations

    location prepositions

    np_max-act · gemini-2.0-flash

    The neuron detects spatial relation words—particularly “behind,” “below,” and similar locative prepositions that describe where things are in a scene.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    털
    -0.07
     Ke
    -0.06
     reminis
    -0.06
    illum
    -0.06
    汽车
    -0.06
     etmesi
    -0.06
     Glad
    -0.06
     sağlamak
    -0.06
    Thunk
    -0.06
    rians
    -0.06
    POSITIVE LOGITS
    нице
    0.07
     above
    0.06
    ,length
    0.06
     PreparedStatement
    0.06
     ниже
    0.06
     وهو
    0.06
     behind
    0.06
    [first
    0.06
    ASC
    0.06
    nette
    0.06
    Act Density 0.019%

    No Known Activations