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

    Neuronpedia

    INDEX
    Explanations

    references to learning from past experiences or mistakes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ieri
    -0.18
    ismatch
    -0.16
    uji
    -0.15
    avic
    -0.15
    acades
    -0.15
     nutrit
    -0.15
    utches
    -0.14
    amilia
    -0.14
    .opendaylight
    -0.14
    enade
    -0.14
    POSITIVE LOGITS
     lessons
    0.81
     Lessons
    0.72
     lesson
    0.71
    lessons
    0.68
     Lesson
    0.65
    Lesson
    0.57
    lesson
    0.57
     learn
    0.54
     learns
    0.49
     learned
    0.49
    Act Density 0.303%

    No Known Activations