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

    Neuronpedia

    INDEX
    Explanations

    makes

    np_max-act · gemini-2.0-flash

    This neuron detects the “What makes …” construction in question headings, firing strongly on the verb “makes” (and the small following word) when a title begins “What makes…”.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     arreglo
    -0.07
    DataTask
    -0.06
     circus
    -0.06
    Bean
    -0.06
    ısını
    -0.06
    \helpers
    -0.06
     nightly
    -0.06
    �
    -0.06
    _merged
    -0.06
     prive
    -0.06
    POSITIVE LOGITS
     REP
    0.07
    บ
    0.07
    кою
    0.07
     CMP
    0.07
    ména
    0.06
    内
    0.06
     zem
    0.06
     ответствен
    0.06
     Eb
    0.06
     Fem
    0.06
    Act Density 0.021%

    No Known Activations