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

    Neuronpedia

    INDEX
    Explanations

    shadows and depth

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Temps
    0.45
     Suicide
    0.44
     tempList
    0.42
    temps
    0.40
     idioms
    0.40
     LCCN
    0.39
     densities
    0.39
    Tmp
    0.39
    UpdateChoice
    0.39
     comuns
    0.38
    POSITIVE LOGITS
     shadow
    0.91
    Shadow
    0.86
    shadow
    0.86
     Shadow
    0.82
    立体
    0.80
     depth
    0.78
     shadows
    0.77
     shadowing
    0.77
     embossed
    0.76
     shadowed
    0.75
    Act Density 0.005%

    No Known Activations