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

    Neuronpedia

    INDEX
    Explanations

    actions that involve physical or emotional interaction with others

    oai_token-act-pair · gpt-3.5-turbo

    actions related to emotional expressions and decision-making

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    umn
    -0.64
    roma
    -0.62
    MSN
    -0.60
    uum
    -0.60
     Extrem
    -0.60
    utenberg
    -0.59
     ][
    -0.59
    >>\
    -0.58
    olon
    -0.57
    UGC
    -0.57
    POSITIVE LOGITS
     accordingly
    0.99
     alike
    0.88
     oneself
    0.84
     yourself
    0.83
     ourselves
    0.81
     yourselves
    0.79
     versa
    0.78
     them
    0.76
    ings
    0.74
    ables
    0.73
    Act Density 0.393%

    No Known Activations