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

    Neuronpedia

    INDEX
    Explanations

    attends to full sentences ending with periods from the beginning of the sentence

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.14
    1:0.13
    2:0.38
    3:0.08
    4:0.08
    5:0.02
    6:0.05
    7:0.08
    Negative Logits
     ourselves
    -0.32
    abestanden
    -0.30
    GETHER
    -0.28
    DECREF
    -0.28
    themselves
    -0.28
     Segura
    -0.28
    
    -0.27
    ESSO
    -0.27
     Autorizaciones
    -0.27
    werf
    -0.27
    POSITIVE LOGITS
    contentLoaded
    0.39
     snippetHide
    0.35
     HasFactory
    0.33
    readFileSync
    0.32
    elemField
    0.32
    IUrlHelper
    0.32
    icza
    0.32
    errHandler
    0.31
    avyzd
    0.31
     województ
    0.31
    Act Density 0.222%

    No Known Activations