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

    Neuronpedia

    INDEX
    Explanations

    comfort despite trade-offs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Deterministic
    0.83
    deterministic
    0.79
     deterministic
    0.78
    混乱
    0.74
     crackers
    0.72
    recursion
    0.70
     কৃত
    0.69
     tjän
    0.69
    agr
    0.68
     દર
    0.67
    POSITIVE LOGITS
     comfort
    2.25
     Comfort
    2.13
    Comfort
    2.10
     cushioning
    2.03
     comfortable
    2.01
     cushioned
    1.99
    comfort
    1.96
     Comfortable
    1.87
     комфор
    1.87
     comfy
    1.86
    Act Density 0.263%

    No Known Activations