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

    Neuronpedia

    INDEX
    Explanations

    essay titles

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    兎
    0.94
     व्हाट
    0.90
    oks
    0.78
    Ads
    0.78
    Что
    0.77
    मल
    0.76
     सोते
    0.75
    苓
    0.75
    చిత
    0.75
     oks
    0.74
    POSITIVE LOGITS
    ↵↵↵↵↵↵↵↵↵↵↵
    0.69
    ↵↵↵↵↵↵↵↵↵
    0.64
     vaulted
    0.63
     Castro
    0.62
    的很
    0.61
    ↵↵↵↵↵↵↵↵
    0.61
     진행
    0.60
     interpol
    0.59
     qui
    0.58
     Zeeman
    0.57
    Act Density 0.011%

    No Known Activations