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

    Neuronpedia

    INDEX
    Explanations

    asking for examples or further details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     compatibility
    0.81
     cinema
    0.73
     malicious
    0.69
     models
    0.68
     animation
    0.68
     board
    0.66
     but
    0.66
     fragile
    0.65
     cour
    0.65
     magnetic
    0.65
    POSITIVE LOGITS
    References
    1.60
    Tags
    1.56
    Keywords
    1.54
    Copyright
    1.50
    Source
    1.44
    Keyword
    1.40
    <eos>
    1.34
    Disclaimer
    1.34
    Thank
    1.31
    Answer
    1.31
    Act Density 0.193%

    No Known Activations