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

    Neuronpedia

    INDEX
    Explanations

    phrases that indicate conversation or dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    -0.94
     للمعارف
    -0.93
    SBATCH
    -0.86
     ddelweddau
    -0.86
    ſelves
    -0.85
    neſs
    -0.85
    >");
    -0.82
    yelitis
    -0.82
    '):
    -0.81
    ]--;
    -0.81
    POSITIVE LOGITS
    So
    0.88
     So
    0.84
    Итак
    0.58
     what
    0.57
    so
    0.56
    EndContext
    0.56
    ,
    0.53
    o
    0.53
    2
    0.53
     
    0.52
    Act Density 0.121%

    No Known Activations