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

    Neuronpedia

    INDEX
    Explanations

    say "were"

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    was
    0.96
    Was
    0.94
     was
    0.81
     Was
    0.80
    ிருக்கிறது
    0.79
     ఉంటుంది
    0.79
     ਹੈ
    0.78
     ہے۔
    0.78
     இருக்கிறது
    0.77
    పోయింది
    0.76
    POSITIVE LOGITS
     were
    2.73
    were
    2.30
     weren
    2.28
    Were
    2.17
     WERE
    2.13
     đều
    2.09
     Were
    2.09
     erano
    1.95
     eivät
    1.87
     étaient
    1.82
    Act Density 0.131%

    No Known Activations