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

    Neuronpedia

    INDEX
    Explanations

    capitalized instances of the word "penalty"

    oai_token-act-pair · gpt-3.5-turbo

    references to penalties and their severity in a regulatory or sports context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rina
    -0.79
    atters
    -0.74
    itect
    -0.74
    yll
    -0.74
    oscope
    -0.73
    bian
    -0.73
    bits
    -0.71
    pocket
    -0.71
    geist
    -0.69
    åĮ
    -0.68
    POSITIVE LOGITS
     penalty
    1.09
     penalties
    1.08
     levied
    1.00
     imposed
    0.88
     sanction
    0.87
     incurred
    0.86
     punishment
    0.85
     Penalty
    0.84
     harshly
    0.80
     punishments
    0.80
    Act Density 0.020%

    No Known Activations