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

    Neuronpedia

    INDEX
    Explanations

    warm contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     định
    0.38
     lực
    0.38
    તાઓ
    0.37
    <unused82>
    0.36
    ależ
    0.35
    ician
    0.35
     Donaldson
    0.35
     предоставляет
    0.35
    উন্ড
    0.35
    쉴
    0.34
    POSITIVE LOGITS
     warm
    0.83
     warms
    0.82
    warm
    0.79
     warming
    0.78
    暖
    0.74
     warme
    0.73
     temperatures
    0.73
     temperature
    0.72
    temperature
    0.71
     Warm
    0.71
    Act Density 0.048%

    No Known Activations