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

    Neuronpedia

    INDEX
    Explanations

    creating risks and vulnerabilities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    这一
    0.51
    嘏
    0.48
     стиль
    0.45
    襞
    0.45
    Dialogue
    0.43
    参见
    0.43
     változat
    0.43
    Эти
    0.42
    Purpose
    0.41
    Identifying
    0.41
    POSITIVE LOGITS
     temperature
    0.54
     propellant
    0.48
     risks
    0.46
     baking
    0.45
     temperatures
    0.45
     voltage
    0.45
     parameters
    0.44
     hairdryer
    0.44
     caused
    0.44
     heating
    0.44
    Act Density 0.005%

    No Known Activations