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

    Neuronpedia

    INDEX
    Explanations

    provided

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     retreat
    -0.08
     reinforce
    -0.07
     vanish
    -0.07
     Salv
    -0.07
     настоя
    -0.07
     dend
    -0.07
    enqueue
    -0.07
     тор
    -0.07
     agon
    -0.07
     bisexual
    -0.07
    POSITIVE LOGITS
    珉
    0.08
     industry
    0.07
    LOUD
    0.07
    诗
    0.07
    CHandle
    0.07
     point
    0.07
    粮
    0.07
     jm
    0.07
    🕐
    0.07
    BLEM
    0.07
    Act Density 0.001%

    No Known Activations