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

    Neuronpedia

    INDEX
    Explanations

    use followed by specific techniques

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Using
    0.73
     Using
    0.64
    Having
    0.62
     pakai
    0.59
    Being
    0.59
    Including
    0.57
    Selain
    0.57
    Usage
    0.57
     USING
    0.56
    Uses
    0.56
    POSITIVE LOGITS
     the
    0.77
     a
    0.74
     techniques
    0.67
     an
    0.67
     only
    0.61
     their
    0.56
     tactics
    0.56
     traditional
    0.54
     something
    0.53
     what
    0.53
    Act Density 0.126%

    No Known Activations