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

    Neuronpedia

    INDEX
    Explanations

    following modal verbs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     the
    0.48
    .
    0.46
     a
    0.45
    。
    0.44
    !
    0.43
    ۔
    0.40
    ."
    0.39
    :
    0.39
    ।
    0.38
    ։
    0.37
    POSITIVE LOGITS
     undergoes
    0.62
    ត្រូវបាន
    0.58
     may
    0.57
     becomes
    0.55
     повинні
    0.54
     should
    0.53
     must
    0.52
     fhould
    0.52
     được
    0.51
     può
    0.49
    Act Density 0.221%

    No Known Activations