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

    Neuronpedia

    INDEX
    Explanations

    couple

    np_max-act · gemini-2.0-flash

    The neuron is triggered by vague quantity expressions or approximate amount words (e.g. “a lot,” “couple of,” “some”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ened
    -0.07
     pravděpodob
    -0.06
    enou
    -0.06
    memo
    -0.06
    кап
    -0.06
     hacked
    -0.06
     luận
    -0.06
    _templates
    -0.06
    ･･
    -0.06
    oned
    -0.06
    POSITIVE LOGITS
    _loop
    0.08
     этого
    0.07
    ुब
    0.07
     born
    0.07
     impression
    0.07
     gotten
    0.07
     accommodate
    0.07
     boys
    0.06
     Croatian
    0.06
     Keeper
    0.06
    Act Density 0.350%

    No Known Activations