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

    Neuronpedia

    INDEX
    Explanations

    sentences with negative constructions using "is"

    oai_token-act-pair · gemini-2.0-flash

    negative phrases or negation

    np_token-act-pair-logits · gpt-4o-mini

    disagreement

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     True
    -0.81
    True
    -0.79
     true
    -0.73
    true
    -0.71
    يكب
    -0.68
    TRUE
    -0.63
    PreferredItem
    -0.62
    BufferException
    -0.60
     TRUE
    -0.60
     False
    -0.59
    POSITIVE LOGITS
    __*/
    0.53
    amaño
    0.46
    magini
    0.46
    ()];
    0.46
     cara
    0.45
    #
    0.45
     getCode
    0.45
    longer
    0.44
     استنادى
    0.44
    }`).
    0.43
    Act Density 3.808%

    No Known Activations