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

    Neuronpedia

    INDEX
    Explanations

    attends to specific tokens related to statistical tests from closing parentheses in later tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.08
    2:0.07
    3:0.51
    4:0.07
    5:0.03
    6:0.06
    7:0.07
    Negative Logits
     незавершена
    -0.54
    complexContent
    -0.53
    DebuggerNonUser
    -0.46
    ReusableCell
    -0.44
     ExecuteAsync
    -0.44
     الحره
    -0.44
    NameInMap
    -0.43
    tangentMode
    -0.42
    tonsoft
    -0.42
     pageContext
    -0.42
    POSITIVE LOGITS
     noastre
    0.41
     cardiaque
    0.40
     intitulée
    0.39
     nucléaire
    0.38
     lumineuse
    0.37
     asupra
    0.37
     frequenza
    0.37
    seamnă
    0.36
     mijne
    0.36
     bouteille
    0.35
    Act Density 1.181%

    No Known Activations