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

    Neuronpedia

    INDEX
    Explanations

    terms for specific words or concepts in different languages

    oai_token-act-pair · gpt-3.5-turbo

    terms that involve definitions or explanations of concepts, particularly those specifying what something is or refers to

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    idav
    -0.81
    icka
    -0.74
    choes
    -0.71
    owsky
    -0.69
    cles
    -0.68
    EED
    -0.68
    vez
    -0.68
    Dash
    -0.66
     supplemented
    -0.65
    ©¶æ¥µ
    -0.64
    POSITIVE LOGITS
    bidden
    0.83
    oman
    0.78
    */(
    0.78
     initials
    0.74
     messenger
    0.72
     noun
    0.68
     insults
    0.66
     Interior
    0.62
     pronounced
    0.62
     loosely
    0.62
    Act Density 0.079%

    No Known Activations