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

    Neuronpedia

    INDEX
    Explanations

    attends to mathematical or technical notation from several preceding tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.18
    2:0.17
    3:0.07
    4:0.18
    5:0.07
    6:0.09
    7:0.07
    Negative Logits
    endphp
    -0.47
     Мексичка
    -0.42
    ftagPool
    -0.42
    UVWXYZ
    -0.42
    queryInterface
    -0.41
     <=",
    -0.38
    ]<<"
    -0.36
    حف
    -0.35
    ]}>
    -0.35
     økonom
    -0.35
    POSITIVE LOGITS
     all
    0.27
     hal
    0.26
    iastes
    0.25
     absol
    0.24
    all
    0.24
     tir
    0.23
     aller
    0.22
    WindowConstants
    0.22
    AutoScaleMode
    0.22
    DebuggerNonUser
    0.22
    Act Density 0.307%

    No Known Activations