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

    Neuronpedia

    INDEX
    Explanations

    numeric targets or goals

    oai_token-act-pair · gpt-3.5-turbo

    references to goals, benchmarks, and thresholds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     specialize
    -0.66
    çīĪ
    -0.64
     diffuse
    -0.62
    opoly
    -0.62
    âĹ¼
    -0.62
     unfocused
    -0.59
    chrome
    -0.59
     incorpor
    -0.59
    opol
    -0.59
    Wing
    -0.58
    POSITIVE LOGITS
     achievable
    0.83
     threshold
    0.82
     thresholds
    0.81
    posts
    0.79
     (>
    0.77
     achieved
    0.76
     reached
    0.74
     attained
    0.73
     (<
    0.72
     (~
    0.71
    Act Density 0.244%

    No Known Activations