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

    Neuronpedia

    INDEX
    Explanations

    converting, writing, combining, positioning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     jurisprudence
    0.29
     consideration
    0.28
    אם
    0.27
     blir
    0.27
     agir
    0.27
     BEING
    0.27
     attentions
    0.27
     stays
    0.27
     pertence
    0.26
     делается
    0.26
    POSITIVE LOGITS
     accurately
    0.47
     convert
    0.47
     safely
    0.42
     efficiently
    0.42
     accommodate
    0.41
     reliably
    0.41
     successfully
    0.40
     enable
    0.39
     assign
    0.39
     translate
    0.38
    Act Density 0.138%

    No Known Activations