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

    Neuronpedia

    INDEX
    Explanations

    references to conditional or pending situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hai
    -0.15
    mey
    -0.15
     immediately
    -0.14
    ryn
    -0.14
     already
    -0.14
     Gill
    -0.14
    ucs
    -0.14
    _compile
    -0.14
    prech
    -0.14
     sometimes
    -0.14
    POSITIVE LOGITS
     fully
    0.23
    /if
    0.19
     xong
    0.18
     complete
    0.17
     finishes
    0.17
     finish
    0.16
     finished
    0.16
     Fully
    0.16
    Fully
    0.16
     completion
    0.15
    Act Density 0.171%

    No Known Activations