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

    Neuronpedia

    INDEX
    Explanations

    reducing and eliminating negatives

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    isChecked
    0.76
     sonuc
    0.75
     लापता
    0.73
    ])
    0.72
    連續
    0.71
    middels
    0.71
     Capabilities
    0.71
     مختلفة
    0.70
     đủ
    0.70
    不同的
    0.70
    POSITIVE LOGITS
     unnecessary
    1.81
     clutter
    1.79
     distractions
    1.71
     harmful
    1.63
     hassles
    1.54
     stigma
    1.52
     burden
    1.51
     burdensome
    1.50
     extraneous
    1.50
     wasteful
    1.46
    Act Density 0.491%

    No Known Activations