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

    Neuronpedia

    INDEX
    Explanations

    negative advice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     may
    1.01
     might
    1.00
    要注意
    0.96
     Depending
    0.93
     But
    0.92
     Peut
    0.92
     depending
    0.91
     BUT
    0.88
     mogą
    0.87
    Depending
    0.86
    POSITIVE LOGITS
     erroneous
    1.01
     fraude
    0.95
    otherwise
    0.91
     hypocrisy
    0.85
     substandard
    0.85
     nefarious
    0.85
     barbaric
    0.84
     OTHERWISE
    0.84
     incompetence
    0.84
     adverse
    0.83
    Act Density 0.113%

    No Known Activations