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

    Neuronpedia

    INDEX
    Explanations

    code and technical questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     internet
    0.85
     receiving
    0.83
     glaring
    0.81
     retweet
    0.79
     instilled
    0.79
     intentionally
    0.78
     negligent
    0.78
     DP
    0.78
     negligently
    0.77
     purposely
    0.77
    POSITIVE LOGITS
    <end_of_turn>
    1.16
    In
    1.01
     In
    0.83
    Alternative
    0.79
    Solution
    0.77
    Problem
    0.76
    Documentation
    0.76
     Giải
    0.75
    sympy
    0.73
    Expo
    0.73
    Act Density 0.108%

    No Known Activations