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

    Neuronpedia

    INDEX
    Explanations

    can navigate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     summertime
    0.44
     periodically
    0.42
     marketable
    0.41
     landowners
    0.39
     gasoline
    0.39
    抜け
    0.38
     springtime
    0.38
     durable
    0.37
     profitable
    0.37
     coast
    0.37
    POSITIVE LOGITS
     تھی۔
    0.54
     تھی
    0.52
     تھا۔
    0.52
     cependant
    0.50
    Namun
    0.50
    ا۔
    0.48
     однако
    0.47
     Tuttavia
    0.47
     تھے۔
    0.46
     Однако
    0.45
    Act Density 0.003%

    No Known Activations