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

    Neuronpedia

    INDEX
    Explanations

    any specific aspect

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     plans
    0.92
     intentions
    0.90
     intent
    0.78
     Intent
    0.78
     Plans
    0.78
    intent
    0.77
     missing
    0.75
     이름을
    0.72
    یں۔
    0.72
    Plans
    0.72
    POSITIVE LOGITS
     аспек
    0.80
     aspect
    0.78
    றும்
    0.76
     Gospod
    0.75
    rofoam
    0.73
     aspek
    0.72
    subsec
    0.72
     základě
    0.71
     suivants
    0.69
     Aspects
    0.69
    Act Density 0.062%

    No Known Activations