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

    Neuronpedia

    INDEX
    Explanations

    references to "cut-off" values and thresholds

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    meanor
    -0.44
    alleri
    -0.43
     surla
    -0.43
    Lond
    -0.41
    DEP
    -0.41
    -0.41
     voud
    -0.41
     attends
    -0.40
     Lond
    -0.39
    tara
    -0.39
    POSITIVE LOGITS
     cutoff
    1.28
    cutoff
    1.05
    utoff
    1.03
    ConstraintMaker
    0.72
     truncated
    0.47
     truncation
    0.43
    DBNull
    0.42
     corte
    0.42
     cut
    0.41
     cutout
    0.40
    Act Density 0.193%

    No Known Activations