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

    Neuronpedia

    INDEX
    Explanations

    specific criminal acts or accusations

    oai_token-act-pair · gpt-3.5-turbo

    punctuation marks indicating the end of sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    stall
    -0.83
    plet
    -0.79
    ikuman
    -0.77
    izont
    -0.77
    gment
    -0.75
     mosqu
    -0.71
     transitions
    -0.71
    emort
    -0.70
    poral
    -0.69
     forgot
    -0.69
    POSITIVE LOGITS
     Critics
    1.25
     Others
    1.10
     Lawyers
    1.08
    Critics
    1.07
     Officials
    1.02
     Meanwhile
    1.00
     Supporters
    0.99
     Experts
    0.99
     Specifically
    0.99
     Concern
    0.98
    Act Density 0.391%

    No Known Activations