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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating errors or issues

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating problems or failures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ilege
    -0.70
    æĢ
    -0.66
     choice
    -0.66
    odd
    -0.64
    ile
    -0.62
    uncture
    -0.62
     reclaimed
    -0.60
    ortment
    -0.60
    pron
    -0.59
     pride
    -0.58
    POSITIVE LOGITS
     smoothly
    0.80
     Seym
    0.78
     unnoticed
    0.75
     havoc
    0.72
     miser
    0.70
    vas
    0.68
     onstage
    0.66
    ikarp
    0.63
     belie
    0.62
     Train
    0.62
    Act Density 0.055%

    No Known Activations