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

    Neuronpedia

    INDEX
    Explanations

    valid

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     valid
    -0.84
     Trained
    -0.79
    Trained
    -0.70
     obvious
    -0.64
     válido
    -0.61
    valid
    -0.56
     válida
    -0.55
     trained
    -0.55
    trained
    -0.54
     Valid
    -0.51
    POSITIVE LOGITS
    ness
    1.03
    ly
    0.92
    less
    0.73
    nesses
    0.72
    mate
    0.67
    ments
    0.65
    nes
    0.65
    iness
    0.65
    lessly
    0.65
     متعلقه
    0.65
    Act Density 0.647%

    No Known Activations