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

    Neuronpedia

    INDEX
    Explanations

    promising future potential

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Typical
    0.69
     Typically
    0.67
    enswert
    0.65
     elapse
    0.65
     usual
    0.65
    させていただ
    0.62
     typical
    0.62
     knowingly
    0.62
    確率
    0.62
    典型
    0.62
    POSITIVE LOGITS
     promise
    2.72
     promises
    2.62
     promising
    2.38
    promises
    2.18
     Promise
    2.16
    promise
    2.16
     promised
    2.09
     Promises
    2.07
     offers
    2.06
     promete
    2.06
    Act Density 0.327%

    No Known Activations