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

    Neuronpedia

    INDEX
    Explanations

    phrases related to persistence and continuation

    oai_token-act-pair · gpt-3.5-turbo

    recurring expressions of exclusivity or singularity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    adelphia
    -0.72
    orno
    -0.65
    ucker
    -0.64
    ás
    -0.61
    Zone
    -0.61
    ansom
    -0.59
    rez
    -0.59
    ajo
    -0.59
    ÃŃn
    -0.58
    ÃŃa
    -0.57
    POSITIVE LOGITS
     alike
    0.98
     thereof
    0.65
     consequently
    0.61
     consequ
    0.61
     depend
    0.60
     comr
    0.58
     accordingly
    0.57
     therefore
    0.55
     versa
    0.55
     importantly
    0.55
    Act Density 0.835%

    No Known Activations