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

    Neuronpedia

    INDEX
    Explanations

    phrases and verbs indicating communication and interaction between people

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.04
    1:0.01
    2:0.04
    3:0.05
    4:0.04
    5:0.11
    6:0.01
    7:0.03
    8:0.36
    9:0.05
    10:0.13
    11:0.06
    Negative Logits
     bon
    -1.74
    lees
    -1.73
     ®
    -1.65
     Surviv
    -1.65
    ��
    -1.63
    kun
    -1.62
     Miko
    -1.59
    cham
    -1.59
    ilee
    -1.55
    ghai
    -1.55
    POSITIVE LOGITS
     replies
    2.50
    Answer
    2.30
     reply
    2.28
     complied
    2.28
     replied
    2.25
     answer
    2.22
     answered
    2.08
     affirmative
    1.98
     hesitated
    1.95
     obliged
    1.94
    Act Density 0.143%

    No Known Activations