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

    Neuronpedia

    INDEX
    Explanations

    do my best to

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     boost
    0.67
     ২৮
    0.66
     year
    0.66
     более
    0.65
     more
    0.64
     half
    0.63
     slightly
    0.63
     nearby
    0.62
     bushels
    0.62
     five
    0.61
    POSITIVE LOGITS
    自己的
    0.79
     swoich
    0.77
    做到
    0.75
     swojego
    0.75
    Selon
    0.70
    Heather
    0.70
    َا
    0.69
    Etiam
    0.69
     etiam
    0.67
    ิน
    0.67
    Act Density 0.001%

    No Known Activations