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

    Neuronpedia

    INDEX
    Explanations

    the letter "t" as a single token

    oai_token-act-pair · gpt-3.5-turbo

    the negative contractions of the verb "to be."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tremend
    -0.80
     indo
    -0.70
     behavi
    -0.68
    Reviewer
    -0.68
     SetTextColor
    -0.66
     Reloaded
    -0.64
     Palestin
    -0.62
    çĭ
    -0.62
     Practices
    -0.61
     packs
    -0.61
    POSITIVE LOGITS
    itles
    1.00
    ween
    0.97
    ional
    0.96
    ople
    0.94
    urtles
    0.93
    otally
    0.92
    itled
    0.92
    apest
    0.91
    roph
    0.90
    ruly
    0.90
    Act Density 0.047%

    No Known Activations