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

    Neuronpedia

    INDEX
    Explanations

    Punctuation in quotations

    np_max-act · gemini-2.0-flash

    This neuron activates on the special placeholder tokens for characters (the “NAME_#” identifiers).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    oter
    -0.07
     whitelist
    -0.06
    露出
    -0.06
    doors
    -0.06
    .Comparator
    -0.06
    _dataset
    -0.06
    Wheel
    -0.06
    _metrics
    -0.06
    _CODE
    -0.06
    .newInstance
    -0.06
    POSITIVE LOGITS
    静
    0.07
     nghiêm
    0.07
    гор
    0.06
     Lun
    0.06
    FINAL
    0.06
    terror
    0.06
     전국
    0.06
     }]↵
    0.06
    ерим
    0.06
    mind
    0.06
    Act Density 0.020%

    No Known Activations