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

    Neuronpedia

    INDEX
    Explanations

    hyperlocal business searches and directories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Techniques
    0.56
     丨
    0.55
    Agreement
    0.55
     Ursula
    0.55
    Jacques
    0.54
    <0xAA>
    0.53
     Mechanisms
    0.53
    }',
    0.52
     Agreement
    0.51
     {
    0.50
    POSITIVE LOGITS
     directory
    0.65
     Yelp
    0.62
     advertisers
    0.61
     advertiser
    0.60
     directories
    0.59
     hyperlocal
    0.56
    廣告
    0.54
    FLU
    0.53
     discriminatory
    0.52
     unique
    0.51
    Act Density 0.051%

    No Known Activations