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

    Neuronpedia

    INDEX
    Explanations

    capital letter "C" followed by a special character

    oai_token-act-pair · gpt-3.5-turbo

    indications of significant changes or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     bear
    -0.81
     endeav
    -0.78
     dece
    -0.78
     hero
    -0.78
     lifes
    -0.78
     invisible
    -0.76
     unstoppable
    -0.76
     orient
    -0.75
     cod
    -0.74
     yes
    -0.74
    POSITIVE LOGITS
    Copyright
    1.62
    Meanwhile
    1.61
    Asked
    1.59
    Related
    1.57
    Contribut
    1.56
    Earlier
    1.55
    Contact
    1.53
    Both
    1.49
    However
    1.48
    Neither
    1.47
    Act Density 0.367%

    No Known Activations