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

    Neuronpedia

    INDEX
    Explanations

    domain names and urls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    也将
    0.58
    يمان
    0.57
    }.
    0.56
    ा
    0.55
    是
    0.55
    ijnlijk
    0.54
     নাও
    0.53
    =.
    0.52
    同事
    0.50
    וחות
    0.50
    POSITIVE LOGITS
    et
    0.84
     in
    0.82
    ad
    0.81
    ра
    0.78
     can
    0.75
     say
    0.72
     avete
    0.72
    es
    0.71
    ag
    0.71
     up
    0.68
    Act Density 0.084%

    No Known Activations