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

    Neuronpedia

    INDEX
    Explanations

    statements of emphasis or confirmation, often starting with "Indeed."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Indeed."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    crow
    -0.65
     Northwest
    -0.63
    chen
    -0.62
    common
    -0.62
     seeds
    -0.62
    chairs
    -0.60
     Swedish
    -0.60
     Crusher
    -0.58
     Columbia
    -0.58
    edu
    -0.58
    POSITIVE LOGITS
    entimes
    0.78
    NESS
    0.76
     guiActiveUn
    0.75
    Indeed
    0.74
     notwithstanding
    0.73
    ional
    0.70
    reys
    0.69
    akedown
    0.69
     Mons
    0.69
    eatures
    0.69
    Act Density 0.010%

    No Known Activations