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

    Neuronpedia

    INDEX
    Explanations

    vary significantly, wildly, dramatically

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    yourself
    1.02
     Yourself
    1.02
     yourself
    0.96
     myself
    0.95
     careful
    0.94
     Careful
    0.87
     cuidadosamente
    0.86
     carefully
    0.86
     முழுவதும்
    0.83
     নিজেই
    0.82
    POSITIVE LOGITS
     unexpectedly
    1.01
     unpredict
    0.97
     organically
    0.92
     itself
    0.92
     naturally
    0.91
     spontaneously
    0.90
     predictably
    0.84
     naturalmente
    0.84
     unchecked
    0.82
     عندي
    0.82
    Act Density 0.553%

    No Known Activations