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

    Neuronpedia

    INDEX
    Explanations

    sentence endings with conjunctions and transition phrases

    oai_token-act-pair · gpt-3.5-turbo

    patterns in punctuation or sentence endings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     prosec
    -0.85
     tyrann
    -0.84
     temptation
    -0.83
     sometimes
    -0.82
     reflex
    -0.82
     bending
    -0.81
     persuasion
    -0.81
     hed
    -0.80
     conject
    -0.80
     favour
    -0.79
    POSITIVE LOGITS
    Tickets
    1.68
    Additionally
    1.62
    According
    1.61
    RELATED
    1.57
    Also
    1.56
    Among
    1.54
    Prior
    1.54
    Along
    1.52
    The
    1.49
    Meanwhile
    1.49
    Act Density 0.329%

    No Known Activations