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

    Neuronpedia

    INDEX
    Explanations

    references to recalls and numbers associated with recalls

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.06
    3:0.08
    4:0.16
    5:0.04
    6:0.05
    7:0.36
    8:0.04
    9:0.03
    10:0.03
    11:0.05
    Negative Logits
    BSD
    -1.49
    ancial
    -1.47
    ocre
    -1.46
    eworks
    -1.43
    SHA
    -1.39
    mble
    -1.39
    iosyncr
    -1.37
    iol
    -1.37
    ymm
    -1.36
    mouth
    -1.36
    POSITIVE LOGITS
     memories
    1.73
     whence
    1.52
     consciousness
    1.42
     recalling
    1.35
     previously
    1.34
     colours
    1.32
     contacts
    1.28
     convict
    1.27
     recy
    1.26
     rele
    1.25
    Act Density 0.003%

    No Known Activations