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

    Neuronpedia

    INDEX
    Explanations

    phrases related to making adjustments or modifications without significant impact or interference

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating actions or processes that can occur without additional requirements or complications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    raq
    -0.96
    geist
    -0.82
    late
    -0.75
    oled
    -0.73
    mun
    -0.71
    aez
    -0.70
     Facts
    -0.70
    soon
    -0.70
    lator
    -0.70
    ounce
    -0.69
    POSITIVE LOGITS
     risking
    1.39
     sacrificing
    1.37
     harming
    1.30
     compromising
    1.30
     needing
    1.25
     worrying
    1.21
     interruption
    1.15
     relying
    1.15
     bothering
    1.14
     disrupting
    1.12
    Act Density 0.049%

    No Known Activations