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

    Neuronpedia

    INDEX
    Explanations

    proper nouns related to organizations or technology

    oai_token-act-pair · gemini-2.0-flash

    international

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     faſt
    -0.90
     houſe
    -0.89
     purpoſe
    -0.88
     ſte
    -0.87
     ſmall
    -0.86
     ſtate
    -0.85
     leſs
    -0.84
     esternos
    -0.83
     Efq
    -0.82
     myſelf
    -0.81
    POSITIVE LOGITS
     too
    0.51
    исленность
    0.50
     what
    0.50
    nelt
    0.50
     a
    0.50
     désir
    0.49
     Clusters
    0.49
     certainement
    0.48
     betreft
    0.48
    breng
    0.48
    Act Density 0.252%

    No Known Activations