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

    Neuronpedia

    INDEX
    Explanations

    phrases or terms indicating causation or reasons

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.09
    3:0.06
    4:0.17
    5:0.01
    6:0.08
    7:0.34
    8:0.02
    9:0.02
    10:0.05
    11:0.07
    Negative Logits
    igun
    -1.82
    agall
    -1.66
    ahime
    -1.61
    league
    -1.51
    epad
    -1.49
    VOL
    -1.46
    daq
    -1.45
    allowed
    -1.43
    habi
    -1.42
    air
    -1.40
    POSITIVE LOGITS
     excellence
    1.64
     Signature
    1.63
     theorem
    1.44
     faulty
    1.43
     rejection
    1.43
     abandonment
    1.43
    PsyNetMessage
    1.42
     rebirth
    1.42
     resemblance
    1.41
     Eternity
    1.40
    Act Density 0.009%

    No Known Activations