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

    Neuronpedia

    INDEX
    Explanations

    various kinds

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     bullies
    0.66
     homeless
    0.63
     broken
    0.63
     mujeres
    0.62
    1
    0.61
     simile
    0.61
     silently
    0.60
    0.59
     blanches
    0.59
     exclusivement
    0.59
    POSITIVE LOGITS
     각종
    0.95
    各种
    0.95
    诸多
    0.91
    各项
    0.88
     বেশকিছু
    0.86
    様々な
    0.86
    各種
    0.85
     allerlei
    0.85
     பல்வேறு
    0.84
     다양한
    0.84
    Act Density 0.381%

    No Known Activations