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

    Neuronpedia

    INDEX
    Explanations

    facing negative conditions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     simples
    0.32
     최대한
    0.30
     meilleur
    0.29
     evitando
    0.29
     cea
    0.29
    Something
    0.28
     haciendo
    0.28
     hopefully
    0.28
     Something
    0.28
     Hopefully
    0.28
    POSITIVE LOGITS
     rampant
    0.79
     widespread
    0.72
     pervasive
    0.69
     increasing
    0.65
     severe
    0.61
     declining
    0.61
     decreased
    0.60
     chronic
    0.60
     increased
    0.58
     persistent
    0.57
    Act Density 0.122%

    No Known Activations