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

    Neuronpedia

    INDEX
    Explanations

    technical terms after specific words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ik
    1.16
    ки
    0.97
    т
    0.97
    ت
    0.96
    ul
    0.93
    can
    0.90
    will
    0.88
    us
    0.86
    та
    0.84
    et
    0.83
    POSITIVE LOGITS
     sensit
    0.61
     categor
    0.54
    AN
    0.52
     pressur
    0.51
     styl
    0.50
    ，
    0.49
     creating
    0.48
     unless
    0.48
    ecie
    0.47
     renormal
    0.47
    Act Density 0.000%

    No Known Activations