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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the concept of change or improvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     tromper
    -0.65
     înc
    -0.64
     NDEBUG
    -0.60
     faisons
    -0.57
     setuptools
    -0.57
     femininas
    -0.56
     démocr
    -0.56
     paradiso
    -0.56
     anahtar
    -0.56
     parvenir
    -0.54
    POSITIVE LOGITS
     worse
    0.91
    principalColumn
    0.72
    AndEndTag
    0.70
     nakalista
    0.68
    worse
    0.64
     hotter
    0.62
     louder
    0.62
     increasingly
    0.61
     progressively
    0.59
     bigger
    0.59
    Act Density 0.074%

    No Known Activations