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

    Neuronpedia

    INDEX
    Explanations

    desired output format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hoping
    0.53
     intending
    0.50
     dimaksud
    0.50
    expected
    0.48
     considerado
    0.48
     expected
    0.47
     chosen
    0.47
     выбран
    0.47
     wishing
    0.47
    Expected
    0.46
    POSITIVE LOGITS
     minimum
    0.61
     outcome
    0.57
     outcomes
    0.52
     norms
    0.48
     maximum
    0.47
     విధంగా
    0.46
     minimale
    0.46
     specifications
    0.46
    minimum
    0.45
     method
    0.44
    Act Density 0.037%

    No Known Activations