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

    Neuronpedia

    INDEX
    Explanations

    phrases that indicate a change in focus or direction

    oai_token-act-pair · gpt-3.5-turbo

    the word "away" in various contexts of shifting focus or direction

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ammy
    -0.76
    milo
    -0.73
     turnover
    -0.69
    ingham
    -0.66
     Organ
    -0.65
     Beans
    -0.63
    iosyn
    -0.63
    tone
    -0.63
     chang
    -0.62
     beans
    -0.62
    POSITIVE LOGITS
     toward
    0.78
     towards
    0.78
    finder
    0.70
    posts
    0.69
    RAG
    0.68
     away
    0.67
    ente
    0.67
     from
    0.67
    awei
    0.66
    rov
    0.66
    Act Density 0.031%

    No Known Activations