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

    Neuronpedia

    INDEX
    Explanations

    attends to numerical tokens from nearby tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.15
    2:0.05
    3:0.04
    4:0.14
    5:0.38
    6:0.05
    7:0.07
    Negative Logits
    SourceChecksum
    -0.36
    gulier
    -0.35
    shed
    -0.34
     SVT
    -0.33
    mybatisplus
    -0.33
     AssemblyTitle
    -0.33
    inigungs
    -0.31
     {{{
    -0.30
    ambiguation
    -0.30
    Overs
    -0.30
    POSITIVE LOGITS
    
    0.27
     الاطلاع
    0.26
    getSeconds
    0.25
    8
    0.24
    出版年
    0.24
    fitrión
    0.23
     χρή
    0.23
     élev
    0.23
    Concurrent
    0.23
     CreateTagHelper
    0.23
    Act Density 0.006%

    No Known Activations