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

    Neuronpedia

    INDEX
    Explanations

    convert to numerical representations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ผ้า
    0.76
     обстоя
    0.72
     регули
    0.67
    anolol
    0.67
     dalších
    0.66
     lớn
    0.65
    ustain
    0.64
     допъл
    0.63
     сю
    0.63
     اہم
    0.63
    POSITIVE LOGITS
     conversion
    1.99
     representation
    1.93
     converted
    1.90
     convert
    1.90
    convert
    1.80
     converting
    1.79
     representations
    1.77
    converted
    1.77
    Converting
    1.76
     conversions
    1.75
    Act Density 0.640%

    No Known Activations