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

    Neuronpedia

    INDEX
    Explanations

    predicting potential issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    。
    0.56
     کی۔
    0.50
    》。
    0.49
    ۔
    0.48
    ).
    0.48
    ।
    0.48
    ។
    0.48
    *.
    0.46
    '.
    0.46
    ].
    0.46
    POSITIVE LOGITS
     may
    0.93
     bukanlah
    0.91
     might
    0.82
     может
    0.76
     môže
    0.74
     seems
    0.71
     může
    0.70
     cannot
    0.69
     potrebbe
    0.67
     isn
    0.66
    Act Density 0.110%

    No Known Activations