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

    Neuronpedia

    INDEX
    Explanations

    best, worst, cheapest, simplest

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     most
    1.85
     наиболее
    1.64
     найбільш
    1.57
    most
    1.49
     MOST
    1.47
    Most
    1.43
     Most
    1.39
    MOST
    1.39
    最も
    1.36
     paling
    1.35
    POSITIVE LOGITS
     terd
    0.83
    abl
    0.79
     Nic
    0.77
     Tou
    0.74
     coars
    0.73
     spars
    0.72
    Nic
    0.72
    ますが
    0.72
     sadd
    0.71
    ear
    0.70
    Act Density 0.210%

    No Known Activations