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

    Neuronpedia

    INDEX
    Explanations

    keeping prejudiced or insulting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     تە
    0.41
     CYAN
    0.39
     difíciles
    0.37
    ́t
    0.37
     প্রতিক
    0.36
    few
    0.36
    難しい
    0.36
     FCA
    0.36
     bot
    0.36
     baton
    0.36
    POSITIVE LOGITS
     Keep
    0.92
    Keep
    0.89
     keep
    0.89
     keeps
    0.84
    keep
    0.80
    保持
    0.77
     Keeps
    0.74
     KEEP
    0.74
     simplify
    0.72
     kept
    0.71
    Act Density 0.000%

    No Known Activations