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

    Neuronpedia

    INDEX
    Explanations

    forms, folds, hours, metrics, strategy

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    يير
    0.56
    سبة
    0.53
    izantes
    0.53
    淍
    0.50
    ético
    0.50
    ਾ
    0.50
    濩
    0.50
    қты
    0.49
    れた
    0.49
    れる
    0.49
    POSITIVE LOGITS
    t
    0.55
    ,
    0.51
     obvi
    0.50
    ↵
    0.50
    I
    0.50
    n
    0.49
     singly
    0.48
     bequem
    0.46
    :
    0.46
    g
    0.45
    Act Density 0.000%

    No Known Activations