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

    Neuronpedia

    INDEX
    Explanations

    dangers, problems, solutions

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     threats
    -0.98
     alternatives
    -0.93
     solutions
    -0.77
     Threats
    -0.70
     closures
    -0.69
     shots
    -0.66
     alternative
    -0.63
    ers
    -0.62
    AndEndTag
    -0.62
     Alternatives
    -0.61
    POSITIVE LOGITS
    iestety
    0.64
     financières
    0.60
     to
    0.59
     générations
    0.59
     senare
    0.58
     âgées
    0.58
     présentés
    0.57
     rayures
    0.56
     vandens
    0.56
     prisonniers
    0.56
    Act Density 0.583%

    No Known Activations