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

    Neuronpedia

    INDEX
    Explanations

    mathematical operators and calculations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     conflict
    0.84
     follow
    0.80
     kant
    0.80
     more
    0.80
     almost
    0.79
    Conflict
    0.79
     back
    0.77
     much
    0.75
     hi
    0.75
    follow
    0.74
    POSITIVE LOGITS
     deltaTime
    1.04
    ogeneous
    0.91
     Насе
    0.90
     numerator
    0.89
     satuan
    0.89
     denominator
    0.88
    ogeneity
    0.87
    obium
    0.86
     বাপ
    0.85
    मता
    0.84
    Act Density 1.032%

    No Known Activations