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

    Neuronpedia

    INDEX
    Explanations

    the more, the better

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     найбільш
    0.41
     並
    0.38
     Proper
    0.38
     Whereas
    0.37
    的最
    0.37
     Superior
    0.36
     možda
    0.35
     більш
    0.35
    の手
    0.35
     ListTile
    0.34
    POSITIVE LOGITS
     desto
    1.05
    수록
    0.70
    越高
    0.60
    越多
    0.56
    thebetter
    0.56
    ยิ่ง
    0.54
     exponentially
    0.52
     chances
    0.51
     sooner
    0.51
     fewer
    0.48
    Act Density 0.023%

    No Known Activations