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

    Neuronpedia

    INDEX
    Explanations

    adjectives and phrases related to negative or dangerous situations

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     increa
    -3.57
     disagre
    -3.54
     reluct
    -3.54
     impra
    -3.48
     affor
    -3.46
     unspeak
    -3.33
     inev
    -3.32
     depic
    -3.26
     gaily
    -3.25
     strick
    -3.23
    POSITIVE LOGITS
    <bos>
    1.63
     items
    0.86
     activities
    0.85
     settings
    0.84
     issues
    0.84
     situations
    0.83
     projects
    0.83
     options
    0.82
     elements
    0.82
     circumstances
    0.82
    Act Density 0.472%

    No Known Activations