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

    Neuronpedia

    INDEX
    Explanations

    preparing for models

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    蚊
    0.46
     สํา
    0.46
    不用
    0.45
     сорти
    0.44
    abinieri
    0.43
    少ない
    0.42
     adjacency
    0.41
     extensible
    0.41
    藪
    0.41
    hObject
    0.40
    POSITIVE LOGITS
     Best
    0.44
    Taste
    0.43
    ঙ্ক
    0.40
    缇
    0.40
     réparation
    0.39
     Tada
    0.39
    Eff
    0.38
    ======
    0.38
     যায়
    0.38
     Opp
    0.37
    Act Density 0.000%

    No Known Activations