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

    Neuronpedia

    INDEX
    Explanations

    this isn't

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Таким
    0.56
     एवं
    0.52
    érieures
    0.51
    Similarly
    0.50
    Thus
    0.50
     therewith
    0.50
     подобных
    0.49
    Suitable
    0.49
    そのような
    0.48
    哪些
    0.48
    POSITIVE LOGITS
     is
    1.57
     isn
    1.36
     happens
    1.16
     seems
    1.14
     represents
    1.14
     doesn
    1.13
     gives
    1.12
     creates
    1.09
     goes
    1.08
     makes
    1.08
    Act Density 0.601%

    No Known Activations