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

    Neuronpedia

    INDEX
    Explanations

    references to "lens" or "prism" as metaphors for perspective

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.08
    3:0.06
    4:0.12
    5:0.02
    6:0.05
    7:0.36
    8:0.04
    9:0.04
    10:0.07
    11:0.08
    Negative Logits
    guard
    -1.53
    keys
    -1.52
    guards
    -1.51
     keyboards
    -1.46
    itars
    -1.43
     barric
    -1.42
    joined
    -1.42
     sacks
    -1.41
     typew
    -1.39
     pian
    -1.38
    POSITIVE LOGITS
     Venture
    1.54
    abal
    1.48
    inka
    1.44
    ��
    1.41
     Inquiry
    1.41
     Higher
    1.40
    gradient
    1.37
     Charity
    1.35
     Profit
    1.33
     Vision
    1.31
    Act Density 0.001%

    No Known Activations