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

    Neuronpedia

    INDEX
    Explanations

    words related to different historical eras

    oai_token-act-pair · gpt-3.5-turbo

    references to different eras in history

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acts
    -0.86
    olulu
    -0.85
    amina
    -0.83
    gettable
    -0.83
    gments
    -0.78
    atem
    -0.77
    ournals
    -0.75
    packing
    -0.75
    task
    -0.73
    application
    -0.71
    POSITIVE LOGITS
     Era
    1.30
     era
    1.24
     epoch
    0.88
     Age
    0.88
     nostalg
    0.87
     Frieza
    0.86
    icals
    0.76
    osaurs
    0.73
     msec
    0.72
     eras
    0.72
    Act Density 0.010%

    No Known Activations