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

    Neuronpedia

    INDEX
    Explanations

    can vary

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    utral
    0.44
    ρισ
    0.42
    kie
    0.38
     এটাই
    0.37
    cta
    0.37
    СС
    0.37
    жень
    0.36
     zagro
    0.36
    requested
    0.35
     solely
    0.35
    POSITIVE LOGITS
     deviate
    0.56
    變
    0.56
     berubah
    0.53
     fluctuate
    0.51
     вари
    0.50
    実際
    0.50
     fluctuates
    0.49
     changed
    0.49
     değiş
    0.48
     変
    0.48
    Act Density 0.003%

    No Known Activations