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

    Neuronpedia

    INDEX
    Explanations

    unusual words or phrases within the text

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    hyde
    -0.80
    street
    -0.79
    zig
    -0.76
     wedge
    -0.76
     Boe
    -0.73
     Xer
    -0.71
    tics
    -0.71
    terson
    -0.70
    jay
    -0.69
     Wy
    -0.69
    POSITIVE LOGITS
    atch
    1.09
    ailability
    0.99
    POSE
    0.94
     Mahjong
    0.92
    ortunately
    0.89
    essim
    0.87
    Applic
    0.86
    itely
    0.84
    ATCH
    0.84
     unbeliev
    0.83
    Act Density 0.618%

    No Known Activations