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

    Neuronpedia

    INDEX
    Explanations

    Play, Jujutsu, Secrets, Kailua

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     음
    0.52
     factored
    0.49
     conducive
    0.48
     integr
    0.47
     metaphorical
    0.47
     subserv
    0.46
    에서의
    0.46
     사용
    0.46
     utensils
    0.46
     FACTORS
    0.46
    POSITIVE LOGITS
    Le
    0.67
    Cedar
    0.66
    Sweden
    0.65
    Spy
    0.65
    Maryland
    0.65
    Al
    0.65
    Arctic
    0.65
    Blue
    0.64
    Willow
    0.64
    Ar
    0.63
    Act Density 0.883%

    No Known Activations