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

    Neuronpedia

    INDEX
    Explanations

    physical dependence tolerance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     нацыяна
    0.85
     מ
    0.84
    pocket
    0.81
     parques
    0.80
     pontos
    0.80
     является
    0.79
     государ
    0.79
    follower
    0.79
    TASK
    0.78
    honey
    0.78
    POSITIVE LOGITS
    <bos>
    0.80
     fewer
    0.77
    द्व
    0.72
    errMsg
    0.67
    ী
    0.67
    .")
    0.66
     unclear
    0.66
    ētu
    0.64
    .}
    0.63
     craindre
    0.63
    Act Density 0.000%

    No Known Activations