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

    Neuronpedia

    INDEX
    Explanations

    safety in numbers

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ял
    -0.09
     Mild
    -0.08
     mild
    -0.08
    ڑ
    -0.08
     Westen
    -0.08
    .Completed
    -0.07
    ريف
    -0.07
    ressa
    -0.07
    تهي
    -0.07
     Beg
    -0.07
    POSITIVE LOGITS
     pooling
    0.14
     economies
    0.13
    Pooling
    0.12
    aggreg
    0.12
     aggregated
    0.12
     aggregation
    0.11
     consolidation
    0.11
     consolidated
    0.11
     bündeln
    0.11
    Aggreg
    0.11
    Act Density 0.052%

    No Known Activations