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

    Neuronpedia

    INDEX
    Explanations

    attends to next-related tokens from sibling-related tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.23
    1:0.23
    2:0.16
    3:0.08
    4:0.08
    5:0.07
    6:0.04
    7:0.08
    Negative Logits
     насељу
    -0.28
    SequentialGroup
    -0.26
    ensore
    -0.26
    TintMode
    -0.25
     ekstra
    -0.24
     Beauchamp
    -0.24
     milioane
    -0.24
     wikipagina
    -0.23
     domani
    -0.23
     milagro
    -0.23
    POSITIVE LOGITS
    awtextra
    0.36
    SourceChecksum
    0.32
     ProtoMessage
    0.31
    AutoScaleMode
    0.31
    RenderAtEndOf
    0.31
    Autoritní
    0.30
     gyhoeddwyd
    0.28
    COUVER
    0.28
    gypti
    0.27
    Cyfeiriadau
    0.27
    Act Density 0.107%

    No Known Activations