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

    Neuronpedia

    INDEX
    Explanations

    common

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     sentences
    0.76
     sentence
    0.75
    sentence
    0.73
    Sentence
    0.72
    シンプルな
    0.71
     words
    0.69
     Worte
    0.69
    Words
    0.68
     kalimat
    0.68
     illusory
    0.67
    POSITIVE LOGITS
     suspects
    0.84
     saying
    0.82
     wisdom
    0.82
     Wisdom
    0.80
     complaining
    0.79
     jokes
    0.78
     Saying
    0.75
     Conventional
    0.74
     joked
    0.73
     saws
    0.73
    Act Density 0.022%

    No Known Activations