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

    Neuronpedia

    INDEX
    Explanations

    criticisms and critiques in text

    oai_token-act-pair · gpt-3.5-turbo

    instances of criticism or negative evaluations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Seah
    -0.73
    aceae
    -0.69
    Yep
    -0.66
    ordial
    -0.63
    Thank
    -0.62
     Congratulations
    -0.61
    ignant
    -0.57
     gasp
    -0.57
    VILLE
    -0.57
    wave
    -0.57
    POSITIVE LOGITS
     overly
    1.01
     excessively
    0.94
     portrayal
    0.93
     excessive
    0.93
     inconsistency
    0.90
     unfairly
    0.89
     lack
    0.89
     unnecessarily
    0.86
     inadequate
    0.85
     insufficient
    0.84
    Act Density 0.585%

    No Known Activations