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

    Neuronpedia

    INDEX
    Explanations

    academic papers

    np_max-act · gemini-2.0-flash

    Neuron 4 fires on header metadata of academic commentaries, especially the “comment on” phrase and associated author–work identifiers.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Ultimately
    -0.07
     повыш
    -0.07
     включа
    -0.06
     enh
    -0.06
     руках
    -0.06
     ق
    -0.06
     saja
    -0.06
     fish
    -0.06
     worship
    -0.06
    ATE
    -0.06
    POSITIVE LOGITS
    	dir
    0.07
    	byte
    0.07
    scroll
    0.07
    ssize
    0.06
    Terminal
    0.06
    atta
    0.06
    HEET
    0.06
    _machine
    0.06
    .selectedIndex
    0.06
     userid
    0.06
    Act Density 0.026%

    No Known Activations