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

    Neuronpedia

    INDEX
    Explanations

    negative connotations associated with Soviet-style elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.03
    1:0.01
    2:0.06
    3:0.06
    4:0.08
    5:0.03
    6:0.20
    7:0.09
    8:0.08
    9:0.05
    10:0.16
    11:0.09
    Negative Logits
    pload
    -1.87
    Effective
    -1.53
     shelves
    -1.50
     Wheels
    -1.48
    Buzz
    -1.44
    ertodd
    -1.43
     Streaming
    -1.43
     regress
    -1.42
     Leilan
    -1.38
     totality
    -1.38
    POSITIVE LOGITS
     tradition
    1.63
    azo
    1.62
    colonial
    1.57
    friendly
    1.52
    cookie
    1.47
    educated
    1.46
    afia
    1.45
    isa
    1.44
     prevailed
    1.44
    gemony
    1.43
    Act Density 0.031%

    No Known Activations