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

    Neuronpedia

    INDEX
    Explanations

    positive change or improvement

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    volving
    0.38
     ספר
    0.37
    itabbo
    0.36
    lhe
    0.35
    Ф
    0.35
     Classification
    0.35
     designations
    0.34
    িনী
    0.33
     указыва
    0.33
    ូប
    0.33
    POSITIVE LOGITS
     mejorar
    0.79
     improve
    0.77
     mejora
    0.76
     melhorar
    0.72
     migliorare
    0.72
     improves
    0.70
     cải
    0.70
     amélior
    0.68
     améliorer
    0.68
     улуч
    0.67
    Act Density 0.115%

    No Known Activations