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

    Neuronpedia

    INDEX
    Explanations

    phrases related to claims or assertions made by multiple parties

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.16
    3:0.23
    4:0.10
    5:0.05
    6:0.16
    7:0.03
    8:0.04
    9:0.04
    10:0.05
    11:0.05
    Negative Logits
    liga
    -1.72
     Sabha
    -1.51
     confir
    -1.45
    reditary
    -1.43
     arch
    -1.35
    >.
    -1.32
     lum
    -1.31
    FTWARE
    -1.30
     additions
    -1.29
    urrence
    -1.26
    POSITIVE LOGITS
     depended
    1.49
    pires
    1.39
     mattered
    1.37
    OULD
    1.35
    Bitcoin
    1.35
    might
    1.30
    would
    1.30
     outwe
    1.28
    Everyone
    1.28
    everyone
    1.27
    Act Density 0.006%

    No Known Activations