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

    Neuronpedia

    INDEX
    Explanations

    phrases related to prediction or expectation

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     dunno
    -0.67
     Goldberg
    -0.60
    */(
    -0.59
     Photograph
    -0.58
     supposedly
    -0.56
     *)
    -0.56
     ostensibly
    -0.55
     zones
    -0.55
    Downloadha
    -0.55
    oros
    -0.54
    POSITIVE LOGITS
     someday
    1.08
     continue
    0.99
     hereafter
    0.97
     future
    0.95
     tomorrow
    0.92
     revisit
    0.86
    future
    0.83
     revis
    0.83
    contin
    0.82
     soon
    0.81
    Act Density 0.894%

    No Known Activations