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

    Neuronpedia

    INDEX
    Explanations

    for graph, tracking, or reporting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    étranger
    0.64
     colère
    0.62
     antecedentes
    0.57
     apuesta
    0.57
    étendue
    0.56
     fleurs
    0.56
     morceaux
    0.54
     ตาย
    0.54
     ಯಾವ
    0.53
     colaborador
    0.53
    POSITIVE LOGITS
     creating
    1.69
     designing
    1.66
     obtaining
    1.57
     managing
    1.55
     selecting
    1.51
     extracting
    1.49
     preparing
    1.48
     handling
    1.47
     constructing
    1.47
     formulating
    1.46
    Act Density 3.382%

    No Known Activations