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

    Neuronpedia

    INDEX
    Explanations

    standardized

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    检查
    -0.08
    ital
    -0.07
     ulcer
    -0.07
     capsule
    -0.07
    팔
    -0.07
    _blocking
    -0.07
    Inc
    -0.07
     incur
    -0.07
    Blocks
    -0.07
     uno
    -0.07
    POSITIVE LOGITS
    atatype
    0.07
    �
    0.07
    otation
    0.07
     Hed
    0.07
     băng
    0.07
     gren
    0.07
     המס
    0.06
    .lower
    0.06
    zeit
    0.06
    -men
    0.06
    Act Density 0.003%

    No Known Activations