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

    Neuronpedia

    INDEX
    Explanations

    the word "very" indicating emphasis or intensity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    awtextra
    -1.04
     autant
    -0.72
     nationales
    -0.68
     τὸν
    -0.65
     maille
    -0.65
     blanches
    -0.65
     culturelles
    -0.64
     religieuses
    -0.63
     vectorielles
    -0.63
     sauvages
    -0.63
    POSITIVE LOGITS
     very
    4.19
    Very
    3.68
    very
    3.66
     Very
    3.59
    VERY
    3.07
     VERY
    3.06
     très
    2.61
     muy
    2.59
     sehr
    2.29
     очень
    2.04
    Act Density 0.079%

    No Known Activations