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

    Neuronpedia

    INDEX
    Explanations

    raw, unfiltered honey

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Dimension
    0.46
     আবার
    0.46
     القلب
    0.42
    Rent
    0.41
    Margins
    0.40
    диви
    0.39
    Diam
    0.39
    টা
    0.38
     없습니다
    0.38
    Roz
    0.37
    POSITIVE LOGITS
    </tr>
    0.48
     Thus
    0.42
     VHS
    0.42
     This
    0.42
    ].
    0.40
     Milk
    0.40
     When
    0.39
     It
    0.39
     How
    0.39
     Just
    0.39
    Act Density 0.005%

    No Known Activations