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

    Neuronpedia

    INDEX
    Explanations

    instances where adjective-adverb pairs are unexpectedly combined

    oai_token-act-pair · gpt-3.5-turbo

    adjectives that describe the quality of something, particularly focusing on poor or well descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    omore
    -0.74
    anos
    -0.73
     impossibility
    -0.72
    oleon
    -0.72
    atorium
    -0.71
     Pione
    -0.67
    anta
    -0.67
    itivity
    -0.67
    ACY
    -0.66
     VIDEOS
    -0.66
    POSITIVE LOGITS
     formatted
    1.38
     calibrated
    1.32
     behaved
    1.32
     constructed
    1.31
     crafted
    1.31
     configured
    1.30
     structured
    1.25
     designed
    1.25
     tuned
    1.22
     organized
    1.22
    Act Density 0.124%

    No Known Activations