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

    Neuronpedia

    INDEX
    Explanations

    Phoenix Claw

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    la
    0.51
    li
    0.49
    air
    0.44
    agic
    0.43
    lla
    0.43
    iation
    0.42
     Besides
    0.42
    atu
    0.41
    ture
    0.41
    ks
    0.41
    POSITIVE LOGITS
     دوباره
    0.54
    改めて
    0.51
     ссылку
    0.49
     tornare
    0.49
     немного
    0.46
     разгово
    0.44
     kembali
    0.44
     солдат
    0.43
    군요
    0.43
     фотоап
    0.43
    Act Density 0.002%

    No Known Activations