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

    Neuronpedia

    INDEX
    Explanations

    comma followed by err

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    error
    0.63
     error
    0.62
    WithError
    0.52
     Error
    0.52
    误差
    0.52
     erreur
    0.51
     ERROR
    0.50
    Error
    0.49
     erreurs
    0.48
    erreur
    0.47
    POSITIVE LOGITS
     suffix
    0.39
     מע
    0.38
     мм
    0.37
    Inte
    0.36
    Selector
    0.35
     prefix
    0.35
     '/'
    0.35
    ള്ള
    0.35
     numele
    0.35
     selector
    0.34
    Act Density 0.000%

    No Known Activations