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

    Neuronpedia

    INDEX
    Explanations

    how to find queries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Particle
    0.40
     Benutz
    0.40
    涎
    0.35
     Particles
    0.34
     Detective
    0.34
     Data
    0.34
     Event
    0.34
    創作
    0.33
     masterful
    0.33
    创作
    0.33
    POSITIVE LOGITS
    How
    0.74
    如何在
    0.69
    how
    0.66
    如何
    0.64
     how
    0.63
     كيفية
    0.62
     How
    0.61
     如何
    0.61
    best
    0.59
    Best
    0.58
    Act Density 0.025%

    No Known Activations