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

    Neuronpedia

    INDEX
    Explanations

    don't, isn't, wasn't, it's

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    വ്
    0.75
    w
    0.75
    går
    0.75
    wap
    0.74
    色が
    0.74
    meye
    0.70
    ları
    0.69
     byla
    0.68
    verkehr
    0.68
    abilir
    0.68
    POSITIVE LOGITS
     unprecedented
    0.90
     reiterate
    0.85
     uncomfortable
    0.83
     uneasy
    0.82
     trembling
    0.82
     то
    0.80
     Timothy
    0.79
     indispensable
    0.78
     reiter
    0.78
     shaking
    0.77
    Act Density 0.118%

    No Known Activations