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

    Neuronpedia

    INDEX
    Explanations

    creating steps or prerequisites

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     often
    0.83
     sometimes
    0.83
     skyrock
    0.79
     thus
    0.77
     also
    0.75
     fought
    0.71
     that
    0.70
     and
    0.70
     typically
    0.70
     therefore
    0.70
    POSITIVE LOGITS
    Install
    0.95
    Calculate
    0.94
    Prepare
    0.92
    Create
    0.91
    Determine
    0.89
     Create
    0.89
     Calculate
    0.88
     Prepare
    0.85
    Identify
    0.85
     Obtain
    0.84
    Act Density 0.223%

    No Known Activations