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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating uncertainty or lack of clarity

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    @#&
    -0.57
    atra
    -0.53
    alez
    -0.51
    visor
    -0.50
    izons
    -0.50
    tes
    -0.49
    INT
    -0.49
    ngth
    -0.48
    gencies
    -0.48
    ailable
    -0.47
    POSITIVE LOGITS
     whether
    0.54
     why
    0.54
     how
    0.51
     chronological
    0.50
    ether
    0.47
     TBD
    0.47
     motive
    0.46
     enough
    0.46
     discern
    0.46
     conclusive
    0.46
    Act Density 11.510%

    No Known Activations