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

    Neuronpedia

    INDEX
    Explanations

    hard/soft

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     юк
    0.65
     오
    0.63
    RxDocument
    0.62
     tenancy
    0.59
     Payload
    0.58
    Picklist
    0.57
    ัย
    0.56
     উঠতে
    0.56
     tenant
    0.55
    portál
    0.55
    POSITIVE LOGITS
     hard
    2.02
     Hard
    2.01
    Hard
    1.88
    hard
    1.79
     HARD
    1.71
     hardness
    1.64
     water
    1.61
    HARD
    1.57
    软
    1.56
     soft
    1.54
    Act Density 0.117%

    No Known Activations