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

    Neuronpedia

    INDEX
    Explanations

    C'est difficile de déterminer le critère de sélection principal pour le neurone 4 étant donné la diversité des activations

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the abbreviation 'D.C.' or references to Washington, D.C

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Maker
    -0.93
    Reviewer
    -0.80
    ãĥĨãĤ£
    -0.72
    ItemTracker
    -0.71
    ãĥŀ
    -0.70
    ãĥĻ
    -0.70
    ãĥij
    -0.68
     mats
    -0.67
    Materials
    -0.65
    éĹĺ
    -0.65
    POSITIVE LOGITS
    urry
    0.82
    .,
    0.78
    rimination
    0.78
    asse
    0.76
    ruck
    0.76
    ember
    0.75
    ollar
    0.74
    uce
    0.71
    hyde
    0.71
    isco
    0.70
    Act Density 0.012%

    No Known Activations