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

    Neuronpedia

    INDEX
    Explanations

    words related to a decrease or decline

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    andum
    -0.69
    aceous
    -0.62
    aneous
    -0.59
    aneously
    -0.59
    achu
    -0.58
     Kenobi
    -0.57
     Kad
    -0.56
     Pros
    -0.56
    acca
    -0.55
    ificent
    -0.54
    POSITIVE LOGITS
    graded
    1.41
    grading
    1.35
    LOAD
    1.32
    stairs
    1.24
    pour
    1.20
    hill
    1.19
    sized
    1.19
    played
    1.18
    grades
    1.15
    vote
    1.14
    Act Density 0.045%

    No Known Activations