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

    Neuronpedia

    INDEX
    Explanations

    code/technical output

    np_max-act · gemini-2.0-flash

    This neuron responds to mentions of the backspace key (e.g. “Backspace,” “back space,” “key”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    EDIATEK
    -0.07
    	cc
    -0.07
    -reg
    -0.07
     substantially
    -0.06
     dict
    -0.06
    	log
    -0.06
    男人
    -0.06
    emente
    -0.06
    大的
    -0.06
    	ptr
    -0.06
    POSITIVE LOGITS
    าหาร
    0.07
    adio
    0.07
    Česk
    0.06
     Offering
    0.06
     Laugh
    0.06
    \-
    0.06
    Nick
    0.06
    ponse
    0.06
    —I
    0.06
    _templates
    0.06
    Act Density 0.203%

    No Known Activations