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

    Neuronpedia

    INDEX
    Explanations

    tokens with the stem `retain`, modified by other descriptive nouns

    oai_token-act-pair · gemini-2.0-flash

    retained/tained

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    jspx
    -0.74
     betweenstory
    -0.70
     Италијани
    -0.66
    Geplaatst
    -0.62
    RunAsync
    -0.60
     disambiguazione
    -0.60
    Hochspringen
    -0.59
    PropertyChanging
    -0.59
    ()<<
    -0.59
     cherchés
    -0.59
    POSITIVE LOGITS
     remaining
    0.88
     remain
    0.84
     jää
    0.76
    留
    0.75
     remains
    0.75
     остались
    0.71
    を残
    0.70
     residual
    0.70
     remained
    0.69
     remainder
    0.68
    Act Density 1.241%

    No Known Activations