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

    Neuronpedia

    INDEX
    Explanations

    safe space

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ר
    0.66
    ма
    0.66
    ור
    0.65
    יל
    0.64
    зи
    0.63
    า
    0.61
    а
    0.59
    ба
    0.59
    א
    0.59
    ор
    0.58
    POSITIVE LOGITS
    0.52
     Se
    0.51
     se
    0.50
     Ch
    0.50
     See
    0.50
     Sh
    0.48
     includes
    0.48
     tiene
    0.48
     As
    0.48
     Sol
    0.48
    Act Density 0.367%

    No Known Activations