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

    Neuronpedia

    INDEX
    Explanations

    linearly increasing growth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    כל
    0.31
    מו
    0.30
    ய
    0.30
    מש
    0.29
    שי
    0.29
    ット
    0.28
     असतो
    0.28
    观念
    0.28
    உலக
    0.28
     знаешь
    0.28
    POSITIVE LOGITS
     HIS
    0.32
     με
    0.31
     Assistant
    0.29
     Walmart
    0.28
     Omicron
    0.28
     aligned
    0.28
     during
    0.27
     protesters
    0.27
     attendees
    0.27
     Jubilee
    0.27
    Act Density 0.340%

    No Known Activations