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

    Neuronpedia

    INDEX
    Explanations

    numeric values or references to achievements or accomplishments

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "record" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    creen
    -0.73
    yg
    -0.72
    adin
    -0.71
     cart
    -0.69
    argon
    -0.66
    ILA
    -0.66
    vette
    -0.64
    nery
    -0.64
    ogie
    -0.64
    dot
    -0.63
    POSITIVE LOGITS
     breaker
    1.12
    breaking
    1.04
    keeping
    0.98
     Breaker
    0.92
     holder
    0.90
     lows
    0.82
     highs
    0.82
     shattering
    0.81
    breaker
    0.78
     achievement
    0.75
    Act Density 0.033%

    No Known Activations