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

    Neuronpedia

    INDEX
    Explanations

    term

    np_max-act · gemini-2.0-flash

    The neuron detects numeric expressions specifying lengths of sentences (e.g. prison terms) in legal text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     treats
    -0.06
     granddaughter
    -0.06
    inta
    -0.06
     ALWAYS
    -0.06
    werk
    -0.06
    िलत
    -0.06
     واحد
    -0.06
    produce
    -0.06
     sufficient
    -0.06
    asal
    -0.06
    POSITIVE LOGITS
    cry
    0.07
     Salesforce
    0.07
    _CLASS
    0.07
    ΟΜ
    0.07
    IGINAL
    0.06
     Timestamp
    0.06
     Executive
    0.06
     универ
    0.06
     Adjustable
    0.06
    
    0.06
    Act Density 0.002%

    No Known Activations