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

    Neuronpedia

    INDEX
    Explanations

    words followed by inappropriate or special characters

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     nurs
    -0.69
     icing
    -0.65
    ensical
    -0.64
    utterstock
    -0.63
     sacrific
    -0.61
     recip
    -0.61
     undergrad
    -0.61
     pulp
    -0.61
    illac
    -0.61
    iewicz
    -0.60
    POSITIVE LOGITS
    ´
    0.84
    rio
    0.83
    Rah
    0.83
    ¯
    0.77
    tri
    0.77
    âĤ¬
    0.76
    til
    0.76
    raid
    0.76
    ready
    0.74
    ¢
    0.74
    Act Density 0.009%

    No Known Activations