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

    Neuronpedia

    INDEX
    Explanations

    multilingual words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵
    0.58
    <h2>
    0.50
    <h3>
    0.49
    <start_of_image>
    0.43
    ↵
    0.42
    ↵↵↵
    0.40
    ↵↵↵↵↵
    0.38
    <h1>
    0.36
    ↵↵↵↵
    0.35
    .”
    0.34
    POSITIVE LOGITS
     menyediakan
    0.56
     初始化
    0.56
     doğrudan
    0.53
     collezione
    0.52
     ensimmä
    0.52
     thisobject
    0.51
    欳
    0.51
     öğrenc
    0.50
    獃
    0.50
    妫
    0.50
    Act Density 1.697%

    No Known Activations