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

    Neuronpedia

    INDEX
    Explanations

    references to correcting or repairing something

    oai_token-act-pair · gemini-2.0-flash

    fix

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     protoimpl
    -0.48
    ItemGroup
    -0.46
    nteral
    -0.45
     filled
    -0.42
    定
    -0.42
    ͘
    -0.42
     thích
    -0.42
     abz
    -0.41
     rộng
    -0.41
    Begriffsklä
    -0.41
    POSITIVE LOGITS
    ]")]
    0.81
    tagHelperRunner
    0.79
    GEBURTSDATUM
    0.76
     itſelf
    0.63
     متعلقه
    0.62
    المناصب
    0.62
    تقاوى
    0.61
    IsContent
    0.59
    StructEnd
    0.59
    TagMode
    0.59
    Act Density 5.436%

    No Known Activations