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

    Neuronpedia

    INDEX
    Explanations

    Creative writing excerpts

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     yu
    -0.07
    _keep
    -0.07
    reste
    -0.06
    rtc
    -0.06
     چشم
    -0.06
     ee
    -0.06
     ec
    -0.06
    ense
    -0.06
     improvised
    -0.06
     cuatro
    -0.06
    POSITIVE LOGITS
    Attention
    0.08
     dưới
    0.06
    Router
    0.06
     tonumber
    0.06
        ↵↵↵
    0.06
    lox
    0.06
     mutated
    0.06
     fortunes
    0.06
    udad
    0.06
    ользоват
    0.06
    Act Density 0.173%

    No Known Activations