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

    Neuronpedia

    INDEX
    Explanations

    compounding frequency and periods

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    й
    0.98
    па
    0.95
    ud
    0.93
    াড়
    0.89
    ţi
    0.89
    šli
    0.88
    arit
    0.88
    দের
    0.87
    pessoas
    0.87
    ំព
    0.86
    POSITIVE LOGITS
     Tick
    0.80
     Stitch
    0.78
     PER
    0.74
     ngo
    0.73
     lL
    0.73
     beso
    0.70
                  
    0.70
     Hod
    0.70
    ن
    0.70
    次数
    0.68
    Act Density 0.005%

    No Known Activations