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

    Neuronpedia

    INDEX
    Explanations

    difference

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     softly
    0.74
     nod
    0.62
     लो
    0.61
     lightly
    0.59
     soft
    0.59
    icons
    0.59
     उच्च
    0.58
     conscious
    0.57
     नरम
    0.57
    と同
    0.56
    POSITIVE LOGITS
     difference
    3.31
    difference
    3.05
     Difference
    3.05
    Difference
    3.00
     diferença
    2.86
     différence
    2.77
     differences
    2.76
     Differences
    2.52
    差
    2.50
    Differences
    2.50
    Act Density 0.652%

    No Known Activations