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

    Neuronpedia

    INDEX
    Explanations

    scientific and technical literature

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     feroit
    -0.97
     étoient
    -0.91
     auroit
    -0.89
     étoit
    -0.87
     avoient
    -0.84
     pouvoit
    -0.82
     meille
    -0.81
     médicas
    -0.80
     déchir
    -0.79
     quelcon
    -0.77
    POSITIVE LOGITS
    ary
    0.60
    MemoryWarning
    0.56
    ative
    0.55
     O
    0.54
     Al
    0.54
     ar
    0.53
    <eos>
    0.52
     W
    0.52
    ated
    0.52
    ers
    0.52
    Act Density 0.074%

    No Known Activations