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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating negation or absence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.03
    1:0.07
    2:0.13
    3:0.08
    4:0.01
    5:0.04
    6:0.10
    7:0.06
    8:0.06
    9:0.11
    10:0.13
    11:0.14
    Negative Logits
    bats
    -1.20
     Researchers
    -1.01
    roup
    -0.99
    fuck
    -0.97
    �
    -0.96
     swung
    -0.96
    antis
    -0.95
    ventures
    -0.95
    Sl
    -0.94
     Talks
    -0.93
    POSITIVE LOGITS
     justification
    1.08
     evidence
    1.07
    ayne
    1.06
     ambiguity
    1.03
     tampering
    1.00
    tera
    1.00
     escaping
    0.98
     satisfactory
    0.98
    ��
    0.98
     inconsistency
    0.96
    Act Density 0.043%

    No Known Activations