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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts or difficult to define

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Andrew
    0.70
     Improved
    0.69
     Improve
    0.66
     Tell
    0.65
    उँ
    0.65
     Reduction
    0.64
     Kevin
    0.63
     Makes
    0.63
     Optimal
    0.63
     vVertex
    0.62
    POSITIVE LOGITS
     intangible
    1.16
     non
    1.12
     perishable
    1.02
     esoteric
    1.00
     complex
    0.98
     ephemeral
    0.97
     miscellaneous
    0.93
    Non
    0.93
    non
    0.93
     mythological
    0.92
    Act Density 0.831%

    No Known Activations