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

    Neuronpedia

    INDEX
    Explanations

    Quotation marks

    np_max-act · gemini-2.0-flash

    This neuron detects explicit Q&A conclusion phrases—especially “the answer is …” or quoted summary answers (e.g. “the answer is ‘yes’” or “‘it depends’”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Modification
    -0.07
    Accessible
    -0.06
    LOYEE
    -0.06
    ्�
    -0.06
    SMS
    -0.06
     mont
    -0.06
    Motor
    -0.06
     MPG
    -0.06
     UIWindow
    -0.06
    -fetch
    -0.06
    POSITIVE LOGITS
    ../../
    0.07
    는다
    0.07
    .Pe
    0.06
     initialise
    0.06
    .assertj
    0.06
     milyar
    0.06
     elim
    0.06
    Resources
    0.06
     fragments
    0.06
    ıydı
    0.06
    Act Density 0.063%

    No Known Activations