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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    tale
    1.96
    till
    1.92
    tone
    1.78
    taker
    1.77
    től
    1.76
    tive
    1.72
    ING
    1.70
    tener
    1.69
    tune
    1.64
    tiden
    1.59
    POSITIVE LOGITS
    на
    2.06
    م
    1.98
    ر
    1.96
    м
    1.89
    ل
    1.84
    nnnn
    1.67
    larda
    1.62
    lardan
    1.62
    י
    1.62
    ized
    1.54
    Act Density 0.290%

    No Known Activations