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

    Neuronpedia

    INDEX
    Explanations

    informative answers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     angla
    0.42
     matemático
    0.42
    卍
    0.41
     CHEMISTRY
    0.41
     protects
    0.41
     आकर्षण
    0.41
     Protect
    0.40
    ifferential
    0.40
    hiddenMap
    0.40
    <unused81>
    0.39
    POSITIVE LOGITS
     comprehensive
    0.56
     informative
    0.43
     Comprehensive
    0.40
     detailed
    0.39
     informed
    0.38
     строк
    0.38
    Comprehensive
    0.38
     or
    0.38
     pre
    0.37
     exhaustive
    0.37
    Act Density 0.001%

    No Known Activations