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

    Neuronpedia

    INDEX
    Explanations

    sets parameters, configurations, or states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     فٹ
    0.42
     quando
    0.38
     मैट्रिक
    0.37
    💵
    0.37
    📓
    0.37
     când
    0.36
     చెర్రీ
    0.36
     lograr
    0.35
    armed
    0.35
    বাচক
    0.35
    POSITIVE LOGITS
    pat
    0.49
    one
    0.42
     PAT
    0.41
     AF
    0.40
     PAH
    0.40
     Patten
    0.40
     ಸ್ವಾ
    0.40
     ext
    0.39
    ceptor
    0.39
     pat
    0.37
    Act Density 0.000%

    No Known Activations