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

    Neuronpedia

    INDEX
    Explanations

    text related to future plans and considerations

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating concerns about focus and distraction in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Defeat
    -0.76
     Vide
    -0.64
    irgin
    -0.64
    ousy
    -0.64
     Pigs
    -0.61
     Prosper
    -0.60
    AAA
    -0.60
     Theft
    -0.59
     attm
    -0.59
    Appearances
    -0.58
    POSITIVE LOGITS
     understandably
    1.26
     wondering
    1.06
     wondered
    0.98
     naturally
    0.96
     inevitably
    0.95
     apprehens
    0.91
     increasingly
    0.91
     suddenly
    0.89
     wonder
    0.87
     beh
    0.87
    Act Density 0.506%

    No Known Activations