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

    Neuronpedia

    INDEX
    Explanations

    attends to special symbols and non-alphanumeric sequences from surrounding alphanumeric tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.09
    2:0.06
    3:0.06
    4:0.18
    5:0.35
    6:0.07
    7:0.07
    Negative Logits
    AsUp
    -0.40
    ReusableCell
    -0.35
    MessageTagHelper
    -0.35
    TemporalType
    -0.33
    osoba
    -0.32
     thủy
    -0.31
    feira
    -0.31
     Mehl
    -0.31
     Astor
    -0.30
    SerializedSize
    -0.30
    POSITIVE LOGITS
    ñoz
    0.28
    ConstraintMaker
    0.27
    ầm
    0.24
     Económica
    0.24
    vább
    0.24
     صوتيه
    0.24
    ğından
    0.23
    stüt
    0.23
     Serrano
    0.23
    iastes
    0.23
    Act Density 0.058%

    No Known Activations