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

    Neuronpedia

    INDEX
    Explanations

    possessive pronouns

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.07
    2:0.09
    3:0.08
    4:0.09
    5:0.06
    6:0.09
    7:0.09
    8:0.08
    9:0.08
    10:0.08
    11:0.07
    Negative Logits
     WARN
    -1.76
    �醒
    -1.63
     unsc
    -1.57
     yields
    -1.56
    GREEN
    -1.52
     rollout
    -1.49
     unaffected
    -1.49
     SOC
    -1.48
     predictable
    -1.46
     yield
    -1.45
    POSITIVE LOGITS
    borgh
    1.93
    odka
    1.65
    aug
    1.64
     Cthulhu
    1.64
     idols
    1.59
    rab
    1.57
    mu
    1.55
     Haunted
    1.54
    thing
    1.53
    aan
    1.53
    Act Density 0.000%

    No Known Activations