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

    Neuronpedia

    INDEX
    Explanations

    positive adjectives and superlative phrases used in reviews.

    oai_token-act-pair · gemini-2.0-flash

    best

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    脚注の使い方
    -0.85
     myſelf
    -0.70
    눠
    -0.69
    ніципа
    -0.68
    PreferredItem
    -0.67
     Riproduzione
    -0.66
    aarrggbb
    -0.66
     whoſe
    -0.66
    Архівовано
    -0.65
     JpaRepository
    -0.65
    POSITIVE LOGITS
     good
    0.75
    good
    0.73
     GOOD
    0.71
     excellent
    0.71
    GOOD
    0.67
     Good
    0.65
    excellent
    0.64
    Good
    0.63
     outstanding
    0.63
     Excellent
    0.62
    Act Density 1.218%

    No Known Activations