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

    Neuronpedia

    INDEX
    Explanations

    recipient understanding

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ছয়
    0.50
     funções
    0.50
    <0xB2>
    0.49
     judge
    0.48
     twor
    0.47
     oč
    0.46
     charg
    0.45
    <0xAE>
    0.45
     océ
    0.45
     province
    0.45
    POSITIVE LOGITS
    ST
    0.51
    pleasant
    0.49
    goods
    0.46
    MRT
    0.46
     itse
    0.45
    Mumbai
    0.44
    ip
    0.44
    UU
    0.44
    이는
    0.44
    ом
    0.43
    Act Density 0.001%

    No Known Activations