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

    Neuronpedia

    INDEX
    Explanations

    statistics comparing averages across different regions or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ZD
    -0.15
    doll
    -0.14
    ara
    -0.14
     Lum
    -0.14
    ĽĪ
    -0.14
    strup
    -0.14
    ando
    -0.14
    oop
    -0.14
    cop
    -0.13
     Source
    -0.13
    POSITIVE LOGITS
     norm
    0.26
    norm
    0.23
     norms
    0.23
     levels
    0.21
    levels
    0.20
     level
    0.20
     baseline
    0.19
     value
    0.19
     Norm
    0.18
    Norm
    0.17
    Act Density 0.108%

    No Known Activations