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

    Neuronpedia

    INDEX
    Explanations

    instructions on punctuation and generation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Who
    0.84
    Who
    0.75
     Wait
    0.75
    Murray
    0.72
     Assignment
    0.72
     Murray
    0.70
     Kad
    0.69
    Assignment
    0.68
    Wh
    0.67
    Birthday
    0.67
    POSITIVE LOGITS
    gruntled
    0.72
    agen
    0.64
    orough
    0.62
    ique
    0.59
    চা
    0.59
    genic
    0.59
    ランキング
    0.58
     luxe
    0.58
     মেয়র
    0.58
     இச
    0.57
    Act Density 0.000%

    No Known Activations