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

    Neuronpedia

    INDEX
    Explanations

    if, then, why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Whilst
    0.47
     whilst
    0.47
    depending
    0.46
    Although
    0.45
     depending
    0.45
     Although
    0.44
     approx
    0.44
     వలన
    0.44
     although
    0.44
     amongst
    0.42
    POSITIVE LOGITS
     moreover
    0.76
     misalnya
    0.59
     bijvoorbeeld
    0.53
     например
    0.53
     meanwhile
    0.52
     naturalmente
    0.50
     beispielsweise
    0.49
    Moreover
    0.48
     dunque
    0.48
     Moreover
    0.46
    Act Density 0.012%

    No Known Activations