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

    Neuronpedia

    INDEX
    Explanations

    capitalized words or phrases

    oai_token-act-pair · gpt-3.5-turbo

    instances of formal or structured expression

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     ability
    -0.85
     sustained
    -0.84
     clin
    -0.82
     barr
    -0.80
     duty
    -0.79
     capacity
    -0.79
     displacement
    -0.78
     cens
    -0.77
     differential
    -0.77
     capability
    -0.77
    POSITIVE LOGITS
    Anyway
    1.93
    Advertisement
    1.92
    Luckily
    1.64
    advertisement
    1.61
    But
    1.58
    RELATED
    1.56
    Thankfully
    1.54
    Fortunately
    1.52
    So
    1.49
    Which
    1.49
    Act Density 0.425%

    No Known Activations