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

    Neuronpedia

    INDEX
    Explanations

    comparative descriptions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that express thoughtful opinions or reflections

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     militar
    -0.63
     distrust
    -0.61
     Grimm
    -0.59
    irt
    -0.59
    Adams
    -0.59
     mistrust
    -0.59
    rus
    -0.59
    roma
    -0.58
    yon
    -0.57
    etheus
    -0.57
    POSITIVE LOGITS
     anymore
    0.77
     articulate
    0.73
     coherent
    0.71
     myself
    0.70
     muster
    0.70
     words
    0.69
     describ
    0.68
     haha
    0.67
     stomach
    0.67
    peat
    0.66
    Act Density 0.297%

    No Known Activations