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

    Neuronpedia

    INDEX
    Explanations

    warm and friendly descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Iceland
    0.52
     Myanmar
    0.49
     typical
    0.49
     Typical
    0.49
     típica
    0.47
     Paragu
    0.46
     Nepal
    0.46
     Indonesia
    0.46
     Perú
    0.46
     characteristic
    0.45
    POSITIVE LOGITS
    Friendly
    0.68
     friendly
    0.67
    friendly
    0.62
     Friendly
    0.59
     warmth
    0.57
     friendliness
    0.57
    warm
    0.54
     freund
    0.54
    温暖
    0.53
    Warm
    0.52
    Act Density 0.017%

    No Known Activations