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

    Neuronpedia

    INDEX
    Explanations

    words and phrases associated with direct actions or relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    omencl
    -0.66
     >=",
    -0.64
     Walkover
    -0.64
    Przypisy
    -0.61
    javadoc
    -0.60
    ModelSerializer
    -0.59
     decorators
    -0.58
     Gazetteer
    -0.58
     initComponents
    -0.58
    uples
    -0.58
    POSITIVE LOGITS
     direct
    1.55
    Direct
    1.54
    DIRECT
    1.44
    direct
    1.44
     DIRECT
    1.40
     Direct
    1.40
     directly
    1.36
     Directly
    1.36
     directa
    1.33
    directly
    1.33
    Act Density 0.154%

    No Known Activations