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

    Neuronpedia

    INDEX
    Explanations

    requests for actions or information through phrases containing verbs in first person perspective and questions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ©¶æ
    -1.01
     sequ
    -0.92
    ensis
    -0.89
    imm
    -0.87
    paio
    -0.86
    displayText
    -0.85
    externalActionCode
    -0.84
    livious
    -0.83
    è£
    -0.82
    adr
    -0.81
    POSITIVE LOGITS
     Want
    1.03
     Yourself
    1.03
    Want
    0.99
     Hate
    0.98
    ãĥĦ
    0.93
    reprene
    0.89
     Qual
    0.83
     Gain
    0.82
     Sick
    0.82
     bookmark
    0.82
    Act Density 2.073%

    No Known Activations