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

    Neuronpedia

    INDEX
    Explanations

    code structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     THEN
    0.80
     then
    0.71
    Then
    0.67
     fanfare
    0.66
     entonces
    0.66
     তখনই
    0.65
     तभी
    0.65
     research
    0.64
     hunting
    0.64
    -">
    0.63
    POSITIVE LOGITS
    <unused1758>
    0.95
     викона
    0.90
    <unused1814>
    0.90
    <unused544>
    0.89
    ্যারি
    0.89
     prepd
    0.88
    諌
    0.87
    ルダー
    0.87
    𝘁
    0.87
     ciclo
    0.86
    Act Density 0.036%

    No Known Activations