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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating something being regulated, controlled, or under certain rules and authorities

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating regulations or requirements related to compliance and oversight

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hots
    -0.81
    horn
    -0.75
    cknowled
    -0.74
    ndra
    -0.73
    started
    -0.72
    cand
    -0.69
    ppa
    -0.68
    lad
    -0.68
    need
    -0.67
    âĤ¬
    -0.67
    POSITIVE LOGITS
     varying
    0.79
     pronounce
    0.73
     ridicule
    0.72
     continual
    0.71
     divine
    0.71
     scrutiny
    0.71
     unforeseen
    0.70
     improv
    0.70
     certain
    0.69
     rigorous
    0.68
    Act Density 0.067%

    No Known Activations