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

    Neuronpedia

    INDEX
    Explanations

    calculate and compute numerical results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     considerados
    0.44
     выбран
    0.42
     considerada
    0.42
    選択
    0.41
     limitado
    0.41
    terminology
    0.40
     recién
    0.40
     चयनित
    0.39
     선택
    0.39
     conquered
    0.39
    POSITIVE LOGITS
     approximate
    0.77
     probabilities
    0.74
     computes
    0.68
     estimate
    0.64
     statistics
    0.63
     Compute
    0.63
     Approximate
    0.62
     compute
    0.61
     coefficients
    0.60
     precise
    0.60
    Act Density 0.103%

    No Known Activations