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

    Neuronpedia

    INDEX
    Explanations

    words related to problems, issues, and troubles

    oai_token-act-pair · gpt-3.5-turbo

    phrases discussing various problems or issues

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    vez
    -0.73
    avement
    -0.72
    mone
    -0.70
    AGES
    -0.69
    ilaterally
    -0.68
    Interstitial
    -0.68
    artney
    -0.64
    Frag
    -0.61
    sense
    -0.60
    Represent
    -0.60
    POSITIVE LOGITS
     downside
    0.90
     takeaway
    0.86
     lesson
    0.79
     caveat
    0.76
     drawback
    0.75
     question
    0.74
     difference
    0.73
     Problem
    0.73
     problem
    0.67
     thing
    0.66
    Act Density 0.188%

    No Known Activations