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

    Neuronpedia

    INDEX
    Explanations

    Thiel, Tears, hyperparameter, pressure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ;")
    0.53
    ٹنے
    0.52
    rizioni
    0.52
    ่
    0.49
    ม
    0.49
     dramas
    0.47
    oje
    0.47
     እንኳ
    0.46
    Ũ
    0.46
     ரயில்
    0.46
    POSITIVE LOGITS
     to
    0.58
    ᱫ
    0.49
     encuentra
    0.49
     tenía
    0.48
    ারে
    0.47
     was
    0.47
     persoane
    0.46
     líqu
    0.45
    ̈́
    0.45
    魷
    0.45
    Act Density 0.000%

    No Known Activations