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

    Neuronpedia

    INDEX
    Explanations

    attends from punctuation tokens marked with periods from arbitrary tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.11
    1:0.13
    2:0.09
    3:0.09
    4:0.08
    5:0.04
    6:0.19
    7:0.24
    Negative Logits
     Marcell
    -0.39
    enumii
    -0.36
     Coss
    -0.34
     Administrativna
    -0.34
     noDo
    -0.33
     Revolutionary
    -0.32
     Mores
    -0.32
     Kamb
    -0.32
    mybatisplus
    -0.31
    enumi
    -0.31
    POSITIVE LOGITS
    MigrationBuilder
    0.37
    Portail
    0.30
     tocó
    0.28
     lò
    0.27
    WritableDatabase
    0.26
    於是
    0.25
     AssemblyTitle
    0.25
     ред
    0.24
     
    0.24
    httphttps
    0.23
    Act Density 0.118%

    No Known Activations