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

    Neuronpedia

    INDEX
    Explanations

    verbs and phrases indicating expectations and comparisons

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to expectations and perceived inadequacies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gul
    -0.65
     sweat
    -0.57
     reluct
    -0.56
     psychology
    -0.54
     parachute
    -0.54
    letters
    -0.53
     Hik
    -0.53
    ãĥĻ
    -0.52
    ronic
    -0.51
     Roose
    -0.50
    POSITIVE LOGITS
     anyways
    0.93
     anyway
    0.90
    !.
    0.85
     because
    0.85
     anymore
    0.82
    .--
    0.78
     ;)
    0.76
     due
    0.75
    .
    0.74
    .#
    0.74
    Act Density 0.146%

    No Known Activations