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

    Neuronpedia

    INDEX
    Explanations

    not

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on negated action phrases expressing persistent inability to heal or escape (e.g. “will not heal,” “won’t go away”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .sys
    -0.07
     bình
    -0.06
    CompatActivity
    -0.06
    бина
    -0.06
    .GetChild
    -0.06
    PLEMENT
    -0.06
    다가
    -0.06
    های
    -0.06
    /dd
    -0.06
     воды
    -0.06
    POSITIVE LOGITS
    -decoration
    0.06
    股票
    0.06
    .pa
    0.06
     spokeswoman
    0.06
     traders
    0.06
    Impact
    0.06
     Suitable
    0.05
     unknow
    0.05
     прест
    0.05
    _predict
    0.05
    Act Density 0.040%

    No Known Activations