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

    Neuronpedia

    INDEX
    Explanations

    likely wobble and drift

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    имость
    0.50
    ور
    0.49
    ่อง
    0.49
     bárm
    0.47
     باید
    0.46
    ید
    0.46
     devrez
    0.46
    های
    0.45
    ेबल
    0.44
    ود
    0.43
    POSITIVE LOGITS
    Alpha
    0.43
     одном
    0.42
     Notably
    0.42
    redu
    0.42
    FProperties
    0.42
    篌
    0.42
    Hay
    0.41
    m
    0.41
    ján
    0.41
     Ghar
    0.41
    Act Density 0.004%

    No Known Activations