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

    Neuronpedia

    INDEX
    Explanations

    warm & welcoming

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ك
    2.72
    ли
    2.13
    ted
    2.02
    lere
    1.99
    lara
    1.87
    ますが
    1.86
    da
    1.85
    lerine
    1.84
    liance
    1.83
    straße
    1.80
    POSITIVE LOGITS
    த்தில்
    1.90
    TE
    1.68
    აცი
    1.52
    aya
    1.48
     capacitors
    1.48
     marshmallows
    1.47
    通过
    1.40
     nuggets
    1.39
     \%$.
    1.36
    สัญญาณ
    1.36
    Act Density 0.020%

    No Known Activations