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

    Neuronpedia

    INDEX
    Explanations

    cons and empire

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    🔍
    0.47
    Focus
    0.43
    ъ
    0.43
    ู้
    0.42
     കാര്യ
    0.42
     याबाबत
    0.42
    Serialization
    0.42
    Gestion
    0.42
    String
    0.42
     باك
    0.42
    POSITIVE LOGITS
     were
    0.66
     was
    0.59
     slightly
    0.57
    were
    0.56
     ছিলেন
    0.54
     blackened
    0.54
     ήταν
    0.52
     étaient
    0.51
     resented
    0.49
     criticized
    0.48
    Act Density 0.005%

    No Known Activations