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

    Neuronpedia

    INDEX
    Explanations

    attends to prepositions from tokens that are relevant to the context or subject of the sentence

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.10
    2:0.07
    3:0.08
    4:0.08
    5:0.06
    6:0.08
    7:0.41
    Negative Logits
     uska
    -0.42
     Roskov
    -0.40
    EndGlobalSection
    -0.38
    ]));
    -0.37
    "):
    -0.35
    AndEndTag
    -0.34
    BrowserModule
    -0.33
    CppCodeGen
    -0.33
     autorytatywna
    -0.33
    unque
    -0.33
    POSITIVE LOGITS
    antaranya
    0.35
     meille
    0.32
     représ
    0.31
     thermally
    0.29
     avoient
    0.29
     nationaux
    0.28
     grijs
    0.28
     rerum
    0.28
     aerial
    0.28
     Tenemos
    0.28
    Act Density 0.001%

    No Known Activations