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

    Neuronpedia

    INDEX
    Explanations

    code optimization

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _L
    -0.08
     Vorr
    -0.07
    jali
    -0.07
    Spider
    -0.07
     moderated
    -0.07
     sar
    -0.07
    شد
    -0.07
    Reveal
    -0.07
     Sar
    -0.07
     commend
    -0.07
    POSITIVE LOGITS
     overhead
    0.15
     unnecessary
    0.15
     unnecessarily
    0.13
     needless
    0.11
     inutile
    0.11
     inefficient
    0.11
     incurred
    0.10
     лиш
    0.10
     everytime
    0.09
     wasted
    0.09
    Act Density 0.014%

    No Known Activations