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

    Neuronpedia

    INDEX
    Explanations

    variable, Beta, Race, backed, contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    казыва
    0.52
    ί
    0.52
     राशन
    0.51
    ная
    0.49
    ы
    0.49
    ाय
    0.49
     Rég
    0.48
     fervor
    0.48
    ی
    0.48
    ात
    0.48
    POSITIVE LOGITS
     omgeving
    0.64
    environments
    0.61
    Cline
    0.59
     environments
    0.57
    沙漠
    0.56
    >
    0.54
    <unused60>
    0.53
    Environ
    0.53
    Bowl
    0.52
    环境保护
    0.52
    Act Density 0.000%

    No Known Activations