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

    Neuronpedia

    INDEX
    Explanations

    conditional checks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    很难
    0.41
     உங்களுக்கு
    0.39
    Europa
    0.38
    ﾐ
    0.34
    ív
    0.34
     ඔබට
    0.34
     проявля
    0.33
     کھیلو
    0.33
     которы
    0.33
     Europa
    0.33
    POSITIVE LOGITS
     (>
    0.55
    かどうか
    0.49
     >=
    0.47
     THEN
    0.46
     (!_
    0.46
     (<
    0.46
     detected
    0.45
     condition
    0.44
    是否
    0.44
     наличие
    0.43
    Act Density 0.207%

    No Known Activations